A collaborative target search method for UAV swarm based on particle swarm algorithm

Abstract

The present invention discloses a method for collaborative target search of an unmanned aerial vehicle (UAV) swarm based on a particle swarm algorithm, which includes distributing and setting multiple UAVs as a particle swarm, performing environmental feedback modeling for the initial positions through the UAVs, and using the current environmental feedback value as the fitness value; initializing the algorithm parameters of the UAV particle swarm and randomly initializing the self-parameters of the UAVs; making the UAVs move to the corresponding positions and calculating the environmental feedback value F(x i ) of each UAV's current position as the fitness value of this position, and recording the movement speed of the UAV as V; and updating the speed and position of the particles through the particle swarm algorithm. The present invention is applicable to an environment with source signals, especially applicable to the positioning tasks of sparse signal sources in a large-scale wide-area environment, such as disaster site rescue, harmful gas leakage source positioning, etc. The present invention is reasonably designed and significantly improves the accuracy and efficiency of target search.

Description

Technical Field

[0001] The present invention relates to the technical field of swarm intelligence and multi-agent target search, and in particular to a drone swarm collaborative target search method based on a particle swarm algorithm. Background Art

[0002] In recent years, the source localization problem has attracted widespread attention and has gradually developed into a research hotspot. This problem assumes that there are multiple signal sources in an unknown environment, and each signal source will continue to emit signals to the outside world. Multiple agents need to collaboratively locate the location of each source based on the signal strength detected in each area, and gather around each signal source at the end of the search to carry out subsequent tasks. Among them, the number and distribution of signal sources are unknown, and the location of the signal source is the point with the maximum signal strength. Therefore, this problem can be transformed into a multi-mode optimization problem, that is, it is necessary to search for multiple global optimal solutions of the objective function at the same time.

[0003] Particle swarm optimization (PSO) is a swarm intelligence algorithm proposed by Kennedy and Eberhart in 1995, inspired by the foraging of bird flocks, to solve global optimization problems. Due to its simple parameters and fast convergence, PSO has become one of the mainstream global optimization techniques and is widely used in various practical problems. However, due to the randomness of particle search, this method is prone to fall into local optimal solutions. Based on this, many scholars have continuously proposed variant algorithms of particle swarms to further improve their optimization performance by balancing the global search and local search capabilities of the swarm.

[0004] There are many similarities and compatibility between the swarm intelligence field and the multi-agent target search field. If the virtual particles in the particle swarm algorithm are replaced by physical agents with certain communication and perception capabilities, the swarm intelligence optimization algorithm can be extended to the real world, that is, to solve the source positioning problem. However, due to the limited signal range of each signal source, especially in the sparse signal source positioning task in a wide-area environment, there is a large area of ​​no signal value in the search environment, that is, the signal strength is 0. When abstracted into a multi-mode optimization problem, there will be a large area of ​​no gradient in the search space, which seriously affects the optimization effect, thereby reducing the accuracy and efficiency of multi-agent search. According to the survey, the current source positioning algorithms all focus on the plane search space, but the height information is actually very useful. With the increase of height, the drone can detect signals in a larger range, but the detected signal value will gradually weaken until it disappears. If the autonomous adjustment of the drone height can be achieved, and the behavior mode from high global rough positioning to low local fine search can emerge autonomously, the source positioning accuracy and efficiency can be significantly improved. Therefore, it is necessary to invent a multi-target search algorithm that considers the autonomous adjustment of the flight height of the drone group. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the problems existing in the above background technology, the present invention is proposed.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a UAV swarm collaborative target search method based on a particle swarm algorithm, which includes distributing and setting multiple UAVs as a particle swarm, modeling the environmental feedback of the initial position through the UAVs, and taking the current environmental feedback value as the fitness value; initializing the algorithm parameters of the UAV particle swarm and randomly initializing the UAV's own parameters; moving the UAVs to the corresponding positions and calculating the environmental feedback value F(x i ) is used as the fitness value of the position, and the speed of the drone is recorded as V; and the speed and position of the particles are updated through the particle swarm algorithm.

[0008] As a preferred solution of the UAV swarm collaborative target search method based on the particle swarm algorithm described in the present invention, the multiple UAVs are all equipped with signal perception sensors, communication and positioning modules, and initial environmental feedback modeling is performed through the signal perception sensors, communication and positioning modules, and the environmental feedback value is calculated as the fitness value of the particle swarm algorithm.

[0009] As a preferred solution of the UAV swarm collaborative target search method based on the particle swarm algorithm of the present invention, the algorithm parameters of the UAV particle swarm are initialized, including the particle swarm size n and the maximum number of iterations G; the parameters of the UAV itself are initialized, including the initial position x of the particle = {x 1 ,x 2 ,…,x n} and speed v = {v 1 ,v 2 ,…,v n}.

[0010] As a preferred solution of the UAV swarm collaborative target search method based on the particle swarm algorithm of the present invention, wherein: the particle swarm algorithm updates the speed and position of the particles:

[0011]

[0012]

[0013] Where d∈{1,2,...,D}, D represents the optimization dimension, ω is called the inertia weight, which is used to measure the impact of the particle velocity in the previous iteration on the current iteration; c 1 、c 2 is a constant acceleration factor, and c 1 =c 2 =2.05; r 1 、r 2 is a D-dimensional random number vector; pbest i represents the historical optimal position of the i-th particle; lbest i represents the historical optimal position of the particle in the neighborhood of the i-th particle; represents the velocity of the ith particle; represents the position of the ith particle.

[0014] As a preferred solution of the UAV swarm collaborative target search method based on the particle swarm algorithm of the present invention, if the current iteration number is greater than the maximum iteration number G, the positioning task is completed, otherwise the UAV particle swarm is moved again to calculate the environmental feedback value F(x i ) is taken as the fitness value of the position, the UAV movement speed is recorded as V, and then the next iteration is carried out.

[0015] As a preferred solution of the UAV swarm collaborative target search method based on particle swarm algorithm described in the present invention, a three-dimensional environmental feedback model is constructed by simulating the influence mechanism of the UAV flight altitude on the signal detection range and strength, which includes: obtaining the current position information of the UAV; calculating the UAV signal detection range; obtaining the signal strength of each position within the signal detection range through a sensor; and taking the maximum signal strength value as the environmental feedback value of the current UAV position.

[0016] As a preferred solution of the UAV group collaborative target search method based on the particle swarm algorithm of the present invention, it is assumed that the signal sources are all located on the ground, that is, the height is 0, and the signal strength gradually weakens with the increase of the distance from them until it disappears. The environmental feedback value at is calculated as follows:

[0017]

[0018]

[0019] Among them, S represents the set of signal source positions, S i represents the position of the i-th signal source, Represents x i The distance to the i-th signal source, Represents x iThe distance from the nearest signal source, W represents the maximum signal strength, R represents the radius of the signal range, and when the distance exceeds the signal range, the signal strength decays to 0.

[0020] As a preferred solution of the UAV group collaborative target search method based on particle swarm algorithm described in the present invention, the signal source gradually weakens with the increase of altitude until it disappears, and the air position The environmental feedback value at is calculated as follows:

[0021]

[0022] in, Indicates ground position The signal strength value at the location where the signal is measured is α, which is the attenuation coefficient and is used to characterize the speed at which the signal strength decays with increasing altitude.

[0023] As a preferred solution of the UAV group collaborative target search method based on particle swarm algorithm described in the present invention, the UAV signal detection range, denoted as θ, will gradually expand. In this method, it is modeled as the current position of the UAV. The side length of the square area centered on is calculated as follows:

[0024]

[0025] Among them, β is used to characterize the impact of altitude on the detection range of drone signals.

[0026] As a preferred solution of the UAV group collaborative target search method based on the particle swarm algorithm described in the present invention, the UAV calculates the maximum value of the signal strength within its signal detection range as the environmental feedback value, which is calculated as follows:

[0027]

[0028] Among them, X i Indicates the current position of the drone, and X indicates any position within the detection range of the drone signal.

[0029] Beneficial effects of the present invention: The present invention is a universal drone swarm target search method, and the particle swarm algorithm based on it can be any particle swarm variant; the search accuracy, anti-interference ability and other performance of this method will also continue to improve with the development of the field of particle swarm intelligent optimization. The present invention is suitable for environments with source signals, especially for positioning tasks of sparse signal sources in large-scale wide-area environments, such as disaster site rescue, positioning of harmful gas leak sources, etc. The present invention is reasonably designed and significantly improves the accuracy and efficiency of target search. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0031] Figure 1 It is a flowchart of the UAV swarm collaborative target search method based on the particle swarm algorithm in the first and second embodiments. DETAILED DESCRIPTION

[0032] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0033] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0034] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0035] Example 1

[0036] Reference Figure 1 , which is the first embodiment of the present invention, and provides a UAV swarm collaborative target search method based on a particle swarm algorithm, comprising the following steps:

[0037] Step 1.1: In the application scenario, multiple drones are deployed in a dispersed manner. Each drone is equipped with a signal sensing sensor, communication and positioning module. The drone swarm is regarded as a particle swarm, and each drone is regarded as a particle.

[0038] Step 1.2: Perform environmental feedback modeling and use the environmental feedback value of each drone’s current position as the fitness value of the particle swarm algorithm.

[0039] Step 1.3: Initialize the particle swarm algorithm parameter settings: particle swarm size n = 50, optimization dimension D = 3, maximum number of iterations G = 100. At the same time, randomly initialize the initial positions of n particles x = {x 1 ,x 2 ,…,x n} and speed v = {v1 ,v 2 ,…,v n}.

[0040] Step 1.4: Move n drones to the positions of their corresponding particles and calculate the environmental feedback value F(x i ) is used as the fitness value of the position, and the UAV movement speed is recorded as V=1.

[0041] Regarding step 1.4, further, the information obtained by each drone is shared to the backend platform.

[0042] Step 1.5: Update the velocity and position of the particles through the particle swarm algorithm. The update formulas are as follows:

[0043]

[0044]

[0045] Where d∈{1,2,...,D}, D represents the optimization dimension, ω is called the inertia weight, which is used to measure the impact of the particle velocity in the previous iteration on the current iteration; c 1 、c 2 is a constant acceleration factor, and c 1 =c 2 =2.05; r 1 、r 2 is a D-dimensional random number vector; pbest i represents the historical optimal position of the i-th particle; lbest i represents the historical optimal position of the particle in the neighborhood of the i-th particle; represents the velocity of the ith particle; represents the position of the ith particle.

[0046] Step 1.6: If the current number of iterations is greater than the maximum number of iterations G, execute step 1.7, otherwise return to step 1.4 for the next iteration.

[0047] Step 1.7: The source localization task is completed.

[0048] Example 2

[0049] Reference Figure 1 , which is the second embodiment of the present invention. This embodiment is based on the previous embodiment and constructs a three-dimensional environmental feedback model by simulating the influence mechanism of the UAV flight altitude on the signal detection range and strength, which includes:

[0050] Get the current location information of the drone, including the plane position and the altitude of the drone;

[0051] Calculate the drone signal detection range. This range varies with altitude. The higher the drone is, the larger the detection range of the plane below the drone is.

[0052] The sensor obtains the signal strength at each location within the signal detection range. The signal strength depends on the position and altitude of the drone and the actual distribution of the target source signal radiation strength.

[0053] The maximum signal strength value is used as the environmental feedback value of the current drone position instead of the signal strength value of the drone's location. This is because the drone can sense a wider area due to its altitude. By using the maximum signal strength value it senses as the environmental feedback value of the current drone position, the entire drone group can be strengthened to search for areas with large signal strength values.

[0054] Assuming that the signal sources are all located on the ground, that is, the height is 0, the signal strength gradually weakens as the distance increases until it disappears. The environmental feedback value at is calculated as follows:

[0055]

[0056]

[0057] Among them, S represents the set of signal source positions, S i represents the position of the i-th signal source, Represents x i The distance to the i-th signal source, Represents x i The distance from the nearest signal source, W represents the maximum signal strength, R represents the radius of the signal range, and when the distance exceeds the signal range, the signal strength decays to 0.

[0058] As the signal source increases in altitude, the signal strength will gradually weaken until it disappears. The environmental feedback value at is calculated as follows:

[0059]

[0060] in, Indicates ground position The signal strength value at the location where the signal is measured is α, which is the attenuation coefficient and is used to characterize the speed at which the signal strength decays with increasing altitude.

[0061] The detection range of the drone signal, denoted as θ, will gradually expand. In this method, it is modeled as The side length of the square area centered on is calculated as follows:

[0062]

[0063] Among them, β is used to characterize the impact of altitude on the detection range of drone signals.

[0064] The drone will calculate the maximum value of the signal strength within its signal detection range as the environmental feedback value, which is calculated as follows:

[0065]

[0066] Among them, X i Indicates the current position of the drone, and X indicates any position within the detection range of the drone signal.

[0067] The present invention is a universal target search method for drone swarms. In the proposed framework, the particle swarm algorithm based on it can be any particle swarm variant. For the first time, the flight altitude of the intelligent agent is considered in the source positioning algorithm, and a three-dimensional environmental feedback model is constructed. On this basis, the particle swarm algorithm is used as a cluster self-organizing collaborative control strategy, and the drone swarm can autonomously emerge from the global rough positioning at high altitude to the local fine search at low altitude. And the search accuracy, anti-interference ability and other performance of this method will also continue to improve with the development of the field of particle swarm intelligent optimization. The drone swarm implementing this method can effectively complete the source positioning task through collaboration, and has broad application prospects.

[0068] Verification: Simulation experiments were carried out in three complex environments: a small-scale search environment, a large-scale sparse signal source environment, and a large-scale dense signal source environment. In each case, the classical SPSO (D. Bratton and J. Kennedy, "Defining a standard for particle swarm optimization," in Proc. IEEE Swarm Intelligence Symposium, pp. 120–127, 2007.), LIPSO (BY Qu, PN Suganthan, and S. Das, "A distance-based locally informed particle swarm model for multimodal optimization," IEEE Transactions on Evolutionary Computation, vol. 17, no. 3, pp. 387–402, 2013.) and the newer HRTPSO (Z.-G. Chen, Z.-H. Zhan, D. Liu, S. Kwong and J. Zhang, "Particle Swarm Optimization with Hybrid Ring Topology for Multimodal Optimization Problems," 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 2044-2049, 2020.) to deploy the self-organizing collaborative control strategy of the drone swarm to complete the target search task. The environmental parameter settings are shown in Table 1:

[0069] Table 1 Multi-type simulation environment parameter settings

[0070]

[0071]

[0072] Table 2 Algorithm comparison experimental data

[0073]

[0074] The number of drones is set to n=50, and the overall situation is shown in Table 2, where A represents the search accuracy, Δ represents the convergence distance, SPSO, LIPSO, and HRTPSO represent the three particle swarm variant algorithms themselves, and SPSO*, LIPSO*, and HRTPSO* represent the three particle swarm variant algorithms after deploying our model.

[0075] Compared with the direct application of SPSO, after adopting the method of the present invention based on SPSO, the search accuracy in each environment of the drone swarm is increased by an average of 9.08%, and the convergence distance is reduced by an average of 47.16%. Compared with the direct application of LIPSO, after adopting the method of the present invention based on LIPSO, the search accuracy in each environment of the drone swarm is increased by an average of 102.58%, and the convergence distance is reduced by an average of 7.40%. Compared with the direct application of HRTPSO, after adopting the method of the present invention based on HRTPSO, the search accuracy in each environment of the drone swarm is increased by an average of 2.63%, and the convergence distance is reduced by an average of 42.08%. The experimental results show that after deploying our model, the search performance and convergence performance of the drone cluster are significantly improved, especially in the task of sparse signal source positioning in a large-scale wide-area environment.

[0076] Importantly, it should be noted that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are only exemplary. Although only a few embodiments are described in detail in this disclosure, it should be readily understood by those who refer to this disclosure that many modifications are possible (e.g., the size, scale, structure, shape and proportion of various elements, and parameter values ​​(e.g., temperature, pressure, etc.), installation arrangement, use of materials, color, directional changes, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in the application. For example, the element shown as integrally formed can be composed of multiple parts or elements, the position of the element can be inverted or otherwise changed, and the nature or number or position of the discrete element can be changed or changed. Therefore, all such modifications are intended to be included in the scope of the present invention. The order or sequence of any process or method steps can be changed or reordered according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure of performing the function described herein, and is not only structurally equivalent but also equivalent structure. Without departing from the scope of the present invention, other replacements, modifications, changes and omissions can be made in the design, operating conditions and arrangement of the exemplary embodiments. Therefore, the invention is not limited to a specific embodiment, but extends to numerous modifications still falling within the scope of the appended claims.

[0077] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention) may not be described.

[0078] It will be appreciated that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but will be a routine task of design, fabrication, and production for those of ordinary skill having the benefit of this disclosure without undue experimentation.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A UAV swarm collaborative target search method based on particle swarm algorithm, Features: include, Multiple drones are distributed and set up as a particle swarm. The environmental feedback modeling of the initial position is performed through the drones, and the current environmental feedback value is used as the fitness value; The plurality of drones are all equipped with signal perception sensors, communication and positioning modules, and the initial environment feedback modeling is constructed through the signal perception sensors, communication and positioning modules, and the environment feedback value is calculated as the fitness value of the particle swarm algorithm; By simulating the impact of the UAV's flight altitude on the signal detection range and strength, a three-dimensional environmental feedback model is constructed, including: Get the current location information of the drone; Calculate the drone signal detection range; Acquire the signal strength at each location within the signal detection range through the sensor; The maximum signal strength value is used as the environmental feedback value of the current drone position; Assuming that the signal sources are all located on the ground, that is, the height is 0, the signal strength gradually weakens as the distance increases until it disappears. The environmental feedback value at is calculated as follows: Among them, S represents the set of signal source positions, S i represents the position of the i-th signal source, Represents x i The distance to the i-th signal source, Represents x i The distance from the nearest signal source, W represents the maximum signal strength, R represents the radius of the signal range, when the distance exceeds the signal range, the signal strength decays to 0; As the signal source increases in altitude, the signal strength will gradually weaken until it disappears. The environmental feedback value at is calculated as follows: in, Indicates ground position The signal strength value at , α is the attenuation coefficient, which is used to characterize the speed at which the signal strength decays with increasing altitude; Initialize the algorithm parameters of the drone particle swarm and randomly initialize the drone's own parameters; Move the drone to the corresponding position and calculate the environmental feedback value F(x i ) as the fitness value of the position, the UAV movement speed is recorded as V; and, The particle speed and position are updated by particle swarm algorithm.

2. The UAV swarm collaborative target search method based on particle swarm algorithm as claimed in claim 1, Features: Initialize the algorithm parameters of the drone particle swarm, including the particle swarm size n and the maximum number of iterations G; initialize the drone's own parameters, including the initial position of the particle x = {x 1 ,x 2 ,…,x n } and speed v = {v 1 ,v 2 ,…,v n }.

3. The UAV swarm collaborative target search method based on particle swarm algorithm as claimed in claim 2, Features: The particle swarm algorithm updates the velocity and position of particles: Where d∈{1,2,...,D}, D represents the optimization dimension, ω is called the inertia weight, which is used to measure the impact of the particle velocity in the previous iteration on the current iteration; c 1 、c 2 is a constant acceleration factor, and c 1 =c 2 =2.05; r 1 、r 2 is a D-dimensional random number vector; pbest i represents the historical optimal position of the i-th particle; lbest i represents the historical optimal position of the particle in the neighborhood of the i-th particle; represents the velocity of the ith particle; represents the position of the ith particle.

4. The UAV swarm collaborative target search method based on particle swarm algorithm as claimed in claim 3, Features: If the current iteration number is greater than the maximum iteration number G, the positioning task is completed, otherwise the drone particle swarm is moved again to calculate the environmental feedback value F(x i ) is taken as the fitness value of the position, the UAV movement speed is recorded as V, and then the next iteration is carried out.

5. The UAV swarm collaborative target search method based on particle swarm algorithm as claimed in claim 1, Features: The detection range of the drone signal, denoted as θ, will gradually expand. In this method, it is modeled as The side length of the square area centered on is calculated as follows: Among them, β is used to characterize the impact of altitude on the detection range of drone signals.

6. The UAV swarm collaborative target search method based on particle swarm algorithm as claimed in claim 5, Features: The drone will calculate the maximum value of the signal strength within its signal detection range as the environmental feedback value, which is calculated as follows: Among them, X i represents the current position of the UAV, and X represents any position within the UAV signal detection range.

Images