AUV path planning method based on hybrid symbiotic organism search algorithm

By using the combination of hybrid symbiotic search algorithm and particle swarm algorithm in AUV path planning, the problem of local optimization and insufficient stability is solved, more efficient path planning is achieved, and the algorithm's global exploration ability and local development ability are enhanced. , significantly improves the performance and efficiency of path planning.

CN118938969APending Publication Date: 2024-11-12GUANGDONG POLYTECHNIC NORMAL UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410996871.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Existing AUV path planning algorithms are prone to falling into local optimization and insufficient stability in complex marine environments, and fail to make full use of population diversity information, resulting in low path planning efficiency and high computational complexity.

Method used

Using a path planning method based on a hybrid symbiotic search algorithm, the algorithm is designed to enhance the global exploration of the algorithm by constructing the objective function, initializing population parameters and individual paths, combining the iterative update mechanism of the particle swarm algorithm. Capability and local development capabilities.

Benefits of technology

It significantly improves the performance and efficiency of path planning, enhances the robustness and stability of the algorithm, and can converge to the global optimal solution more quickly, avoiding local optimal traps.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938969A_ABST
    Figure CN118938969A_ABST
Patent Text Reader

Abstract

The invention discloses an AUV path planning method based on a hybrid symbiotic organism search algorithm. Comprising the following steps: constructing a complete map model; initializing algorithm parameters by using a map model; taking the starting point and the terminal point as end points, determining a linear path, segmenting the path, constructing a new coordinate system on the segmented path, performing population initialization on the new coordinate system, determining the path of each individual in the population, and randomly initializing the speed of the individual; calculating fitness values of all individuals in the population, recording optimal fitness values of the population and the individuals and corresponding optimal paths, and updating parameters; iteratively updating individual paths; and judging whether an iteration termination condition is met or not, and if the termination condition is not met, continuing iteration until a global optimal path is finally output. According to the method, a new information interaction mechanism and a new benefit factor calculation mode are designed, the global exploration capability and the local development capability of the algorithm are effectively enhanced, and the performance and the efficiency of path planning are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of path planning, and more specifically, relates to an AUV path planning method based on a hybrid symbiotic organism search algorithm. Background Art

[0002] Autonomous Underwater Vehicles (AUVs) have the advantages of strong maneuverability, high versatility, and high intelligence. In recent years, AUVs have shown great application potential in the field of marine resource exploration, covering multiple key areas such as marine surveys, environmental monitoring, data collection, aquaculture, and industrial equipment testing. However, when performing tasks in complex marine environments, AUVs will be affected by factors such as weak communication, low visibility, and biological infestation. In order to ensure that AUVs can complete their tasks safely and efficiently underwater, autonomous path planning is required based on the environment.

[0003] The goal of AUV path planning is to find a path from the starting point to the end point so that it can meet the mission requirements while maximizing energy conservation and avoiding obstacles. Common path planning algorithms include the A* algorithm, the dynamic window method DWA, the fast random tree algorithm RRT, and the probability graph-based algorithm PRM. However, in the actual application of underwater navigation, the objective function and related constraints are very complex. When using these methods to perform path planning under multiple constraints, there are often problems such as high time complexity and easy to fall into local optimal values, and it is impossible to find the global optimal solution. The AUV path planning problem can be regarded as a non-deterministic NP-hard problem. The optimization method based on swarm intelligence algorithm has great potential in dealing with NP-hard problems.

[0004] Swarm intelligence algorithms have the advantages of simple structure, easy implementation, and good effect. At present, many researchers have applied this algorithm to AUV path planning scenarios. The paper "Dynamic path planning of a three-dimensional underwater AUV based on an adaptive genetic algorithm [J]. Ocean Engineering, 2022, 263: 112421." aims to solve the problems of low global path quality and poor dynamic obstacle avoidance performance in underwater three-dimensional AUV autonomous path planning. It proposes a genetic algorithm with adaptive genetic operators and improves the calculation method of the resultant force direction by using the space vector method. However, although the adaptive genetic operator improves efficiency, it still faces problems of local optimality and algorithm stability when dealing with complex marine environments.

[0005] The invention patent with the prior art publication number CN116107322A proposes a three-dimensional path planning method for an autonomous underwater robot. The path is planned by constructing an energy consumption optimal fitness function. When the path planning is not optimal, the advantaged group and the disadvantaged group are optimized and updated by the particle swarm algorithm according to the feedback of the energy consumption optimal fitness function, and the path planning is then carried out according to the energy consumption optimal fitness function until the optimal path is planned. There is no information interaction between individuals in the population of this scheme, and there is a problem of failing to fully utilize the individual position information.

[0006] In summary, adjustment strategies based on parameters or the introduction of certain functions can improve the performance of the algorithm. However, since the diversity information of the population is not fully utilized, there may be a risk of improper parameter adjustment. A method that can fully utilize the diversity of the population is needed. Swarm intelligence algorithms often have the problem of being prone to local optimality and insufficient stability. Many algorithm improvement methods in existing work make the algorithm more complicated. How to propose a method to improve the performance of the algorithm and improve the defects of the algorithm without increasing the computational complexity of the algorithm is the problem that the present invention is committed to solving and the goal to achieve. Summary of the invention

[0007] In order to overcome the problems in the prior art of easily falling into local optimum, insufficient algorithm stability, and failure to fully utilize individual position information, the present invention provides an AUV path planning method based on a hybrid symbiotic organism search algorithm.

[0008] The present invention provides an AUV path planning method based on a hybrid symbiotic organism search algorithm, comprising the following steps:

[0009] Determine the map range of the area that needs path planning, and build a complete map model based on threat area information and terrain obstacle information, define the starting point and end point; use the map model to initialize algorithm parameters and population parameters;

[0010] Taking the starting point and the end point as endpoints, a straight line path is determined, the path is segmented, a new coordinate system is constructed on the segmented path, the population is initialized on the new coordinate system, the path of each individual in the population is determined, and the speed of the individual is randomly initialized;

[0011] Constructing an objective function according to the first preset formula, using the objective function to calculate the fitness values ​​of all individuals in the population, recording the optimal fitness values ​​of the population and individuals and the corresponding optimal paths, and updating the relevant parameters of the algorithm according to the second preset formula;

[0012] Iteratively update individual paths according to preset algorithm rules;

[0013] Determine whether the iteration termination condition is met. If not, return to the fitness value of all individuals in the calculation population and continue iterative calculation until the termination condition is finally met, then stop the iteration and output the global optimal path.

[0014] Furthermore, the initialization algorithm parameters and population parameters specifically include: the dimension of the algorithm calculation space, the size of the population, the maximum number of iterations, the minimum and maximum values ​​of the inertia parameters in the algorithm, the minimum and maximum values ​​of the global learning factor and the self-learning factor, the weight coefficient of the fitness function, the global optimal fitness value, the attenuation factor of the adaptive function, and the minimum and maximum values ​​of the speed of individuals in the population.

[0015] Furthermore, the method for determining the path of each individual in the population is to determine a straight line segment with a starting point and an end point as endpoints, divide the connected line segment into multiple equal-length line segments, draw a straight line perpendicular to each line segment, randomly select a point on each straight line as a path point, and connect the starting point, each path point and the end point in sequence to form a complete path.

[0016] Furthermore, the first preset formula for constructing the objective function is:

[0017]

[0018] Among them, μ represents the weight coefficient, and the objective function F cost The constraints include three parts: fuel cost constraint F fuel , Threat Constraint F threat , Pitch Angle Constraint and the deflection angle constraint F θ .

[0019] Furthermore, the calculation formula of the fuel cost constraint is:

[0020]

[0021] Among them, set f fuel =1, length represents the length of the entire path, and the calculation formula is as follows:

[0022]

[0023] Where n is the number of line segments, and the coordinates of path point i on the line segment are (x i ,y i , z i ).

[0024] Furthermore, the threat constraint F threat The calculation formula is:

[0025] Fthreat =F hit +F danger

[0026] Among them, the ocean terrain collision constraint F hit The calculation method is as follows:

[0027]

[0028] Among them, z i represents the Z-axis coordinate of path point i, f(x i ,y i ) indicates the calculation coordinate (x i ,y i ) terrain height, C hit Set to a constant value q;

[0029] In the marine environment, there are dangerous areas other than terrain threats. If the path is in the threat area, the dangerous area constraint F danger The calculation method is as follows:

[0030]

[0031] Among them, the threat cost f of the kth path is danger The calculation formula of (k) is as follows:

[0032]

[0033] Among them, l k represents the length of the kth path, m represents the number of threat areas, represents the distance from the center of the ith threat area to the position of j times the length of the kth segment, r i represents the radius of the ith threat area.

[0034] Furthermore, the pitch angle constraint and the yaw angle constraint F θ The calculation formula is:

[0035]

[0036] Among them, the pitch angle of the path formed by the kth path point to the k+1th path point and the yaw angle θ:

[0037]

[0038] Define the i-th path

[0039]

[0040] Define f of the i-th path θ :

[0041]

[0042] in, is the maximum pitch angle in the path, θ max is the maximum value of the yaw angle in the path.

[0043] Furthermore, the method for recording the optimal fitness value and the corresponding optimal path is:

[0044] If there is an individual fitness value in the population that is better than the global optimal fitness value, then the global optimal fitness value and the global optimal path are updated;

[0045] If the current fitness value of an individual in the population is better than the individual's best fitness value, the individual's best fitness value and the individual's best path are updated.

[0046] Furthermore, the second preset formula for updating the algorithm related parameters is:

[0047]

[0048] c1=c max -(c max -c min )·(1-e -k·t )

[0049] c2=c max -(c max -c min )·(1-e -k·t )

[0050] Where ω is the inertia parameter, c1 is the self-learning factor, c2 is the global learning factor, k is the decay factor, t is the current iteration number, T is the maximum iteration number, ω max and ω min represents the maximum and minimum values ​​of the inertia weight ω, c max and c min represents the maximum and minimum values ​​of the learning factor, and e represents the Euler number.

[0051] Furthermore, the algorithm for iteratively updating the individual paths is:

[0052] If t<3T / 4, the path update method of the search algorithm is as follows:

[0053] Randomly select the i-th and j-th individuals from the population without duplication, and use the following formula to calculate the speed of the two individuals:

[0054]

[0055] Among them, X i represents the path of the current individual i, i,best represents the optimal path of the current individual i, X j represents the path of the current individual j, X j,best represents the optimal path of the current individual j, G best Represents the current global optimal path, r1 and r2 represent random numbers in the range of (0,1). After calculating the speeds of the two individuals, the speeds are used as interaction information to obtain the common vector Mutual_Vector. The calculation formula is as follows:

[0056]

[0057] Use the symbiotic search algorithm to perform collaborative search and update individual paths. The calculation formula is as follows:

[0058]

[0059] in, For the new path of the individual, the benefit factors BF1 and BF2 are calculated using the following formula:

[0060]

[0061] Among them, fitness() represents the function method for obtaining the fitness value of the individual;

[0062] If t ≥ 3T / 4, the path update method of the search algorithm is as follows:

[0063] The entire population is divided into two parts according to the individual fitness value, X i and X j Taken from two different parts, where X i Randomly select from the population with better fitness value, X j The calculation formula for generating new individuals is as follows:

[0064]

[0065] Among them, r3 represents a random number with a value range of (-1,1). After generating a new individual, the fitness value is calculated and compared with the original individual fitness value:

[0066] If the new individual fitness value is worse than the original individual fitness value, the path will not be updated;

[0067] If the fitness value of the new individual is better than the fitness value of the original individual, the parasitic strategy update path is performed to replace the original individual. The calculation formula is as follows:

[0068]

[0069] Among them, the smaller the fitness value, the better the planned path.

[0070] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0071] The present invention designs a new information interaction mechanism and benefit factor calculation method by integrating the symbiotic organism search algorithm (SOS) and the particle swarm algorithm (PSO), which effectively enhances the global exploration ability and local development ability of the algorithm, improves the robustness and stability of the algorithm, and significantly improves the performance and efficiency of path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to make the purpose and technical solution of the present invention clearer, the present invention provides the following drawings and descriptions:

[0073] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0074] Figure 2 A schematic diagram of the path planning segmentation and SD coordinate system provided in an embodiment of the present invention;

[0075] Figure 3 A schematic diagram of AUV environmental threat constraint calculation provided by an embodiment of the present invention;

[0076] Figure 4 A two-dimensional comparison diagram of path planning results of different algorithms provided in an embodiment of the present invention;

[0077] Figure 5 A three-dimensional comparison diagram of path planning results of different algorithms provided in an embodiment of the present invention;

[0078] Figure 6 A schematic diagram comparing changes in fitness values ​​of different algorithms provided in an embodiment of the present invention; DETAILED DESCRIPTION

[0079] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0080] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0081] The present invention provides an AUV path planning method based on a hybrid symbiotic biological search algorithm, such as Figure 1 The figure shows a flow chart of an AUV path planning method based on a hybrid symbiotic organism search algorithm. The specific steps are as follows:

[0082] S1: Determine the map range of the area that needs path planning, and build a complete map model based on the threat area information and terrain obstacle information, define the starting point and end point; use the map model to initialize the algorithm parameters and population parameters;

[0083] More specifically, the algorithm parameters and population parameters initialized in step S1 include: the dimension of the algorithm calculation space, the size of the population, the maximum number of iterations, the minimum and maximum values ​​of the inertia parameters in the algorithm, the minimum and maximum values ​​of the global learning factor and the self-learning factor, the weight coefficient of the fitness function, the global optimal fitness value, the attenuation factor of the adaptive function, and the minimum and maximum values ​​of the speed of individuals in the population.

[0084] In a specific embodiment, map information is first obtained to determine the scope of the path planning area, and then an environmental model is established based on the threat area information.

[0085] The coordinates (x, y, z) in the three-dimensional coordinate system are used to represent the coordinate points in the path. The path planning area of ​​the AUV can be expressed as: {(x, y, z)|x min ≤x≤x max ,y min ≤y≤y max ,z min ≤z≤z max}. Where x min 、x max ,y min ,y max 、z min 、z max Set them to 0m, 1000m, 0m, 1000m, 0m, 1000m respectively. Define S as the starting point (0,0,0), and define D as the end point (1000,1000,1000).

[0086] Initialize the algorithm parameters and population parameters, including the algorithm dimension, population size, maximum number of iterations, minimum and maximum values ​​of the inertia parameters in the algorithm. Minimum and maximum values ​​of the global learning factor and self-learning factor. And the weight coefficient of the fitness function = 0.4. The global optimal fitness value is initialized to positive infinity, and the attenuation factor of the adaptive function k = 0.1.

[0087] S2: Taking the starting point and the end point as endpoints, determine a straight line path, segment the path, construct a new coordinate system on the segmented path, initialize the population on the new coordinate system, determine the path of each individual in the population, and randomly initialize the speed of the individual;

[0088] More specifically, the method for determining the path of each individual in the population in step S2 is to determine a straight line segment with the starting point and the end point as endpoints, divide the connected line segment into multiple equal-length line segments, draw a straight line perpendicular to each line segment, randomly select a point on each straight line as a path point, and connect the starting point, each path point and the end point in sequence to form a complete path. In a specific embodiment, the path is segmented, a new coordinate system is constructed, and the population is initialized.

[0089] First, if Figure 2 As shown, connect the starting point S and the end point D, divide the line segment SD into 11 equal-length line segments through 10 points, and draw 10 straight lines perpendicular to the line segment SD, represented by {L1, L2, L3, …, L k ,…L n}, in each line segment L i Take a path point P i , connecting the starting point, path points and end point in sequence to form a complete path.

[0090] A new SD coordinate system is constructed based on the starting point S(0,0,0) and the end point D(1000,1000,1000). The line segment SD is used as the X axis of the new coordinate system. S becomes the new zero point, recorded as O'. A new X'Y' coordinate system is constructed based on the SD line segment. The angle θ between OX and O'X' is 45°, that is, the XY coordinate system is rotated 45° counterclockwise. Then the path point (x k ,y k ,z k ) in the SD coordinate system The calculation equation is as follows:

[0091]

[0092] The coordinates of the kth path point x(k) can then be calculated using the following formula, where |SD| represents the length of the straight line connecting the starting point S to the end point D:

[0093]

[0094] The position coordinates X of each individual in the population in the X dimension i The position coordinates of n points that are equal to the average segment SD are expressed as X i =(x(1),x(2),…,x(n)). The positions in the Y and Z dimensions are randomly initialized.

[0095] S3: construct an objective function according to the first preset formula, use the objective function to calculate the fitness values ​​of all individuals in the population, record the optimal fitness values ​​of the population and individuals and the corresponding optimal paths, and update the relevant parameters of the algorithm according to the second preset formula;

[0096] More specifically, the first preset formula for constructing the objective function in step S3 is:

[0097]

[0098] Among them, μ represents the weight coefficient, and the objective function F cost The constraints include three parts: fuel cost constraint F fuel , Threat Constraint F threat , Pitch Angle Constraint and the deflection angle constraint F θ .

[0099] More specifically, the calculation formula of the fuel cost constraint is:

[0100]

[0101] Among them, set f fuel =1, length represents the length of the entire path, and the calculation formula is as follows:

[0102]

[0103] Where n is the number of line segments, and the coordinates of path point i on the line segment are (x i ,y i , z i ).

[0104] More specifically, the threat constraint F threat The calculation formula is:

[0105] F threat =F hit +F danger

[0106] Among them, the ocean terrain collision constraint F hit The calculation method is as follows:

[0107]

[0108] Among them, z i represents the Z-axis coordinate of path point i, f(x i ,y i ) indicates the calculated coordinates (x i ,y i ) terrain height, C hit is set to a constant value, which is equal to 10 in a specific embodiment;

[0109] In the marine environment, there are dangerous areas other than terrain threats. If the path is in the threat area, the dangerous area constraint F danger The calculation method is as follows:

[0110]

[0111] Among them, the threat cost f of the kth path danger The calculation formula of (k) is as follows:

[0112]

[0113] Among them, l k represents the length of the kth path, m represents the number of threat areas, represents the distance from the center of the ith threat area to the position of j times the length of the kth segment, r i represents the radius of the ith threat area, such as Figure 3 shown.

[0114] More specifically, the pitch angle constraint and the yaw angle constraint F θ The calculation formula is:

[0115]

[0116] Among them, the pitch angle of the path formed by the kth path point to the k+1th path point and the yaw angle θ:

[0117]

[0118] Define the i-th path

[0119]

[0120] Define f of the i-th path θ :

[0121]

[0122] in, is the maximum pitch angle in the path, θ max is the maximum value of the yaw angle in the path.

[0123] More specifically, the method for recording the optimal fitness value and the corresponding optimal path in step S3 is:

[0124] If there is an individual fitness value in the population that is better than the global optimal fitness value, then the global optimal fitness value and the global optimal path are updated;

[0125] If the current fitness value of an individual in the population is better than the individual's best fitness value, the individual's best fitness value and the individual's best path are updated.

[0126] More specifically, in order to improve the convergence speed of the algorithm, the linear adaptive parameter and the nonlinear adaptive parameter e are introduced. -kt , the second preset formula for updating algorithm related parameters in step S3 is:

[0127]

[0128] c1=c max -(c max -c min )·(1-e -k·t )

[0129] c2=c max -(c max -c min )·(1-e -k·t )

[0130] Where ω is the inertia parameter, c1 is the self-learning factor, c2 is the global learning factor, k is the decay factor, t is the current iteration number, T is the maximum iteration number, ω max and ω min represents the maximum and minimum values ​​of the inertia weight ω, c max and c min represents the maximum and minimum values ​​of the learning factor, and e represents the Euler number.

[0131] S4: Iteratively update individual paths according to preset algorithm rules;

[0132] More specifically, the algorithm for iteratively updating the individual paths in step S4 is:

[0133] If t<3T / 4, the path update method of the search algorithm is as follows:

[0134] Randomly select the i-th and j-th individuals from the population without duplication, and use the following formula to calculate the speed of the two individuals:

[0135]

[0136] Among them, X i represents the path of the current individual i, X i,best represents the optimal path of the current individual i, X j represents the path of the current individual j, X j,best represents the optimal path of the current individual j, G bestRepresents the current global optimal path, r1 and r2 represent random numbers in the range of (0,1). After calculating the speeds of the two individuals, the speeds are used as interaction information to obtain the common vector Mutual_Vector. The calculation formula is as follows:

[0137]

[0138] Use the symbiotic search algorithm to perform collaborative search and update individual paths. The calculation formula is as follows:

[0139]

[0140] in, For the new path of the individual, the benefit factors BF1 and BF2 are calculated using the following formula:

[0141]

[0142] Among them, fitness() represents the function method for obtaining the fitness value of the individual;

[0143] If t ≥ 3T / 4, the path update method of the search algorithm is as follows:

[0144] The entire population is divided into two parts according to the individual fitness value, X i and X j Taken from two different parts, where X i Randomly select from the population with better fitness value, X j The calculation formula for generating new individuals is as follows:

[0145]

[0146] Among them, r3 represents a random number with a value range of (-1,1). After generating a new individual, the fitness value is calculated and compared with the original individual fitness value:

[0147] If the new individual fitness value is worse than the original individual fitness value, the path will not be updated;

[0148] If the fitness value of the new individual is better than the fitness value of the original individual, the parasitic strategy update path is performed to replace the original individual. The calculation formula is as follows:

[0149]

[0150] Among them, the smaller the fitness value, the better the planned path.

[0151] S5: Determine whether the iteration termination condition is met. If not, return to the fitness value of all individuals in the calculation population and continue iterative calculation until the termination condition is finally met, then stop the iteration and output the global optimal path.

[0152] The following is a table of experimental comparison algorithms.

[0153] Table 1

[0154]

[0155]

[0156] like Figure 4 The following is a two-dimensional comparison diagram of the path planning results of different algorithms. Figure 5 The figure shows a three-dimensional comparison diagram of the path planning results of different algorithms. From the experimental results, it can be seen that the present invention integrates the symbiotic organism search algorithm (SOS) and the particle swarm algorithm (PSO) to design a new information interaction mechanism and benefit factor calculation method, which effectively enhances the global exploration ability and local development ability of the algorithm, and the path obtained by the algorithm is more reasonable.

[0157] like Figure 6 The figure shows a comparison diagram of fitness value changes of different algorithms. From the experimental results, it can be seen that compared with other methods, the present invention can converge to the optimal fitness value more quickly, which significantly improves the performance and efficiency of path planning.

[0158] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A path planning method for AUV based on hybrid symbiotic biological search algorithm, It is characterized in that The steps include: Determine the map range of the area that needs path planning, build a complete map model based on threat area information and terrain obstacle information, and define the starting point and end point; Use the map model to initialize algorithm parameters and population parameters; Taking the starting point and the end point as endpoints, a straight line path is determined, the path is segmented, a new coordinate system is constructed on the segmented path, the population is initialized on the new coordinate system, the path of each individual in the population is determined, and the speed of the individual is randomly initialized; Constructing an objective function according to the first preset formula, using the objective function to calculate the fitness values ​​of all individuals in the population, recording the optimal fitness values ​​of the population and individuals and the corresponding optimal paths, and updating the relevant parameters of the algorithm according to the second preset formula; Iteratively update individual paths according to preset algorithm rules; Determine whether the iteration termination condition is met. If not, return to the fitness value of all individuals in the calculation population and continue iterative calculation until the termination condition is finally met, then stop the iteration and output the global optimal path.

2. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The initialization algorithm parameters and population parameters specifically include: the dimension of the algorithm calculation space, the size of the population, the maximum number of iterations, the minimum and maximum values ​​of the inertia parameters in the algorithm, the minimum and maximum values ​​of the global learning factor and the self-learning factor, the weight coefficient of the fitness function, the global optimal fitness value, the attenuation factor of the adaptive function, and the minimum and maximum values ​​of the speed of individuals in the population.

3. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The method for determining the path of each individual in the population is to determine a straight line segment with a starting point and an end point as endpoints, divide the connected line segment into multiple equal-length line segments, draw a straight line perpendicular to each line segment, randomly select a point on each straight line as a path point, and connect the starting point, each path point and the end point in sequence to form a complete path.

4. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The first preset formula for constructing the objective function is: Among them, μ represents the weight coefficient, and the objective function F cost The constraints include three parts: fuel cost constraint F fuel , Threat Constraint F threat , Pitch Angle Constraint and the deflection angle constraint F θ .

5. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 4, It is characterized in that The calculation formula of the fuel cost constraint is: Among them, set f fuel =1, length represents the length of the entire path, and the calculation formula is as follows: Where n is the number of line segments, and the coordinates of path point i on the line segment are (x i ,y i , z i ).

6. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 4, It is characterized in that The threat constraint F threat The calculation formula is: F threat =F hit +F danger Among them, the ocean terrain collision constraint F hit The calculation method is as follows: Among them, z i represents the Z-axis coordinate of path point i, f(x i ,y i ) indicates the calculated coordinates (x i ,y i ) terrain height, C hit Set to a constant value q; In the marine environment, there are dangerous areas other than terrain threats. If the path is in the threat area, the dangerous area constraint F danger The calculation method is as follows: Among them, the threat cost f of the kth path is danger The calculation formula of (k) is as follows: Among them, l k represents the length of the kth path, m represents the number of threat areas, represents the distance from the center of the ith threat area to the position of j times the length of the kth segment, r i represents the radius of the ith threat area.

7. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 4, It is characterized in that The pitch angle constraint and the yaw angle constraint F θ The calculation formula is: Among them, the pitch angle of the path formed by the kth path point to the k+1th path point and the yaw angle θ: Define the i-th path Define f of the i-th path θ : in, is the maximum pitch angle in the path, θ max is the maximum value of the yaw angle in the path.

8. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The method for recording the optimal fitness value and the corresponding optimal path is: If there is an individual fitness value in the population that is better than the global optimal fitness value, then the global optimal fitness value and the global optimal path are updated; If the current fitness value of an individual in the population is better than the individual's best fitness value, the individual's best fitness value and the individual's best path are updated.

9. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The second preset formula for the update algorithm related parameters is: c 1 =c max -(c max -c min )·(1-e -k·t ) c 2 =c max -(c max -c min )·(1-e -k·t ) Where ω is the inertia parameter, c 1 is the self-learning factor, c 2 is the global learning factor, k is the decay factor, t is the current number of iterations, T is the maximum number of iterations, ω max and ω min represents the maximum and minimum values ​​of the inertia weight ω, c max and c min represents the maximum and minimum values ​​of the learning factor, and e represents the Euler number.

10. The AUV path planning method based on the hybrid symbiotic organism search algorithm according to claim 1, It is characterized in that The algorithm for iteratively updating individual paths is: If t<3T / 4, the path update method of the search algorithm is as follows: Randomly select the i-th and j-th individuals from the population without duplication, and use the following formula to calculate the speed of the two individuals: Among them, X i represents the path of the current individual i, X i,best represents the optimal path of the current individual i, X j represents the path of the current individual j, X j,best represents the optimal path of the current individual j, G best represents the current global optimal path, r 1 and r 2 Represents a random number with a value range of (0,1). After calculating the speeds of the two individuals, the speeds are used as the mutual information to obtain the common vector Mutual_Vector. The calculation formula is as follows: Use the symbiotic search algorithm to perform collaborative search and update individual paths. The calculation formula is as follows: Among them, X inew , X jnew For the new path of the individual, the benefit factor BF is calculated using the following formula 1 and BF 2 : Among them, fitness() represents the function method for obtaining the fitness value of the individual; If t ≥ 3T / 4, the path update method of the search algorithm is as follows: The entire population is divided into two parts according to the individual fitness value, X i and X j Taken from two different parts, where X i Randomly select from the population with better fitness value, X j The calculation formula for generating new individuals is as follows: X inew =X i (t)+r 3 ·(X best (t)-X j (t)) Among them, r 3 Indicates a random number with a value range of (-1,1). After generating a new individual, calculate the fitness value and compare it with the original individual fitness value: If the new individual fitness value is worse than the original individual fitness value, the path will not be updated; If the fitness value of the new individual is better than the fitness value of the original individual, the parasitic strategy update path is performed to replace the original individual. The calculation formula is as follows: Among them, the smaller the fitness value, the better the planned path.

Citation Information

Cited By

  • Robot autonomous navigation system and method based on the dynamic window approach

    TWI919585B