A mobile robot path planning method based on the mayfly algorithm with dynamic adaptive parameter adjustment

Abstract

The present invention discloses a mobile robot path planning method based on a mayfly algorithm with dynamic adaptive parameter adjustment, comprising the following steps: S1, creating a robot environment map by using a grid method; S2, initializing relevant parameters of the mayfly algorithm with dynamic adaptive parameter adjustment; S3, updating the adaptive parameter values ​​in the algorithm; S4, evaluating individual fitness, finding the individual optimal position and the group optimal position; S5, updating the position speed of male and female mayflies, and evaluating the fitness value; S6, sorting male and female mayflies, and performing male and female mating in sequence; S7, randomly dividing the offspring into male and female; S8, sorting male and female mayflies, retaining the first n excellent individuals, and updating the individual optimal position and the group optimal position; S9, judging whether the iteration stop condition is met, if so, executing S10; otherwise, the number of cycles is increased by 1, and S3 is executed; S10, outputting the saved optimal path. The present invention not only improves the global optimal solution of the algorithm but also improves the convergence speed of the algorithm.

Description

Technical Field

[0001] The invention relates to the technical field of robot path planning, and in particular to a mobile robot path planning method based on an ephemera algorithm with dynamic adaptive parameter adjustment. Background Art

[0002] With the advancement of science and technology, the continuous development of Internet technology, the application of robots has become more and more extensive, and mobile robots have played a huge role in various fields such as production and life. In the use of mobile robots, path planning plays a huge role and is an extremely important part of the realization of mobile robot functions. Path planning refers to the realization of a feasible path from the starting point to the target point by a mobile robot under certain constraints that meets specific optimization criteria. Path planning research has been continuously developing in recent years, and there are many excellent optimization methods, such as the traditional artificial potential field method, A* algorithm, rapid expansion random tree method, and emerging intelligent algorithms such as particle swarm algorithm, ant colony algorithm, genetic algorithm, mayfly algorithm, etc.

[0003] The Mayfly algorithm is a new intelligent optimization algorithm proposed by Konstantinos Zervoudakis in 2020. It simulates the flight behavior and mating process of mayflies, and combines the advantages of swarm intelligence and evolutionary algorithms. It can be considered as an improvement of the particle swarm optimization algorithm, combining the advantages of particle swarm, genetic algorithm and firefly algorithm. However, while the Mayfly algorithm has various advantages, it also contains some disadvantages, such as slow convergence speed, easy to fall into local optimal solution, poor solution accuracy, etc. In response to these shortcomings, many scholars at home and abroad have tried to improve the traditional Mayfly algorithm. Although a large number of simulation results show that some improvement strategies on the basic Mayfly algorithm are feasible and effective, there are still some defects that need to be made up. For example, the imbalance between global search and local search caused by the fixed inertia weight and learning factor makes the optimization ability insufficient in the early stage, which is easy to produce local optimal, and easy to rush out of the optimal area in the later stage, resulting in insufficient accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a mobile robot path planning method based on the mayfly algorithm with dynamic adaptive parameter adjustment, which can overcome the shortcomings of the traditional mayfly algorithm, such as slow convergence speed and easy to fall into local optimum, and not only improves the quality of obtaining the global optimal solution of the robot path planning, but also improves the convergence speed.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A mobile robot path planning method based on the mayfly algorithm with dynamic adaptive parameter adjustment comprises the following steps:

[0007] Step 1: Initialize the environment, digitally encode the motion environment, and map the encoded 0-1 matrix into a grid graph model;

[0008] Step 2: Initialize the fixed parameters of the algorithm: the position x, speed v, number n of male and female mayflies, positive attraction coefficient a3 of female mayflies, visual acuity coefficient β, random walk coefficient fl, maximum speed Velmax, minimum speed Velmin, number of iterations T, and dance coefficient dance;

[0009] Step 3 Update the adaptive parameter values: positive attraction coefficient a1, a2, inertia weight g

[0010] Step 4: Evaluate individual fitness, find the individual optimal pbest and the group optimal position gbest, and calculate the fitness according to formula (1);

[0011]

[0012] Step 5: Update the position and speed of male and female mayflies through formulas (2)-(5) to evaluate fitness;

[0013]

[0014]

[0015]

[0016]

[0017] where r p ,r g is x i and the Cartesian distance between the local optimum and the global optimum, e is a random number between [-1,1], r mf is the Cartesian distance between males and females.

[0018] Step 6: Sort the male and female mayflies according to their fitness from low to high, and mate them in order according to formulas (6) and (7);

[0019] offspring1=L*male+(1-L)*female (6)

[0020] offspring2=L*female+(1-L)*male (7)

[0021] Where male is the male parent, female is the female parent, and L is a random value in [0,1].

[0022] Step 7: randomly divide the offspring into males and females;

[0023] Step 8: Sort the male and female mayflies according to their fitness, retain the top n excellent individuals, and update the optimal positions of the individuals and the group;

[0024] Step 9 determines whether the iteration stop condition is met. If so, execute step 10; otherwise, increase the number of loops by 1 and execute step 3;

[0025] Step 10 outputs the saved optimal path.

[0026] The beneficial effect of the present invention is that the mayfly algorithm is improved based on the dynamic adaptive parameter adjustment strategy, the global optimal solution can be obtained, the efficiency and stability of path planning solution are improved, the balance between local search capability and global search capability is improved through dynamic adaptive parameter adjustment, the execution efficiency of the algorithm is accelerated, and the optimization capability of the algorithm is improved. The simulation experiment results show that the performance of the mayfly algorithm is significantly improved by adopting the novel dynamic adaptive parameter adjustment improvement strategy of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

[0028] Figure 1 Schematic diagram of the raster graph model;

[0029] Figure 2 The optimal paths obtained by the two algorithms;

[0030] Figure 3 Iterative convergence comparison chart of the two algorithms;

[0031] Figure 4 The optimal paths obtained by the two algorithms;

[0032] Figure 5 Comparison chart of iterative convergence of the two algorithms. DETAILED DESCRIPTION

[0033] The purpose of the present invention is to provide a mobile robot path planning method of an improved mayfly algorithm with a dynamic adaptive parameter adjustment strategy. The method improves the shortcomings of the traditional mayfly algorithm, such as slow convergence speed and easy to fall into local optimum, and not only improves the quality of obtaining the global optimal solution of the robot path planning, but also improves the convergence speed.

[0034] In path planning research, environmental modeling is the primary task. So far, there are many related environmental modeling methods applied to path planning problems, such as grid graph method, topology graph method, visibility graph method, etc. Among them, the grid graph method is widely used because of its simple modeling and easy implementation. Therefore, this paper selects the grid graph method as the environmental modeling method. In the grid graph model, blank grids are usually used to represent movable grids, and black grids are used to represent obstacle grids, that is, inaccessible grids. Figure 1Shown is a schematic diagram of a raster image environment.

[0035] In the process of path planning by traditional mayfly algorithm, inertia weight and positive attraction coefficient are usually set to a fixed value, which makes the local search ability unbalanced with the global search ability, and is easy to fall into the local optimum, and also limits the iteration speed. After setting the inertia weight as an adaptive parameter, the particles can be given a larger speed in the early stage of optimization, which can improve the ability to search for the global optimal solution. When approaching the optimal solution in the later stage, the speed is reduced so that the particle speed is not too large and deviates from the optimal position area, missing the global optimal solution and falling into the local optimum. After setting the positive attraction coefficient as an adaptive parameter, the algorithm keeps a balance between local optimization and global optimization, and more experience is obtained from itself in the early stage, which increases the diversity of algorithm results and avoids falling into the local optimum. In the later stage, more experience comes from the group, which increases the accuracy of the results.

[0036] In the improved Mayfly algorithm with dynamic adaptive parameter adjustment strategy, the adaptive parameters are set as shown in formulas (8)-(10)

[0037]

[0038]

[0039]

[0040] The technical solution adopted by the present invention to solve its technical problem is:

[0041] A mobile robot path planning method based on an improved mayfly algorithm with a dynamic adaptive parameter adjustment strategy comprises the following steps:

[0042] Step 1: Initialize the environment, digitally encode the motion environment, and map the encoded 0-1 matrix into a grid graph model;

[0043] Step 2: Initialize the fixed parameters of the algorithm: the position x, speed v, number n of male and female mayflies, positive attraction coefficient a3 of female mayflies, visual acuity coefficient β, random walk coefficient fl, maximum speed Velmax, minimum speed Velmin, number of iterations T, and dance coefficient dance;

[0044] Step 3 Update the adaptive parameter values: positive attraction coefficient a1, a2, inertia weight g

[0045] Step 4: Evaluate individual fitness, find the individual optimal pbest and the group optimal position gbest, and calculate the fitness according to formula (1);

[0046]

[0047] Step 5: Update the position and speed of male and female mayflies through formulas (2)-(5) to evaluate fitness;

[0048]

[0049]

[0050]

[0051]

[0052] where r p ,r g is x i and the Cartesian distance between the local optimum and the global optimum, e is a random number between [-1,1], r mf is the Cartesian distance between males and females.

[0053] Step 6: Sort the male and female mayflies according to their fitness from low to high, and mate them in order according to formulas (6) and (7).

[0054] offspring1=L*male+(1-L)*female (6)

[0055] offspring2=L*female+(1-L)*male (7)

[0056] Where male is the male parent, female is the female parent, and L is a random value in [0,1].

[0057] Step 7: randomly divide the offspring into males and females;

[0058] Step 8: Sort the male and female mayflies according to their fitness, retain the top n excellent individuals, and update the optimal positions of the individuals and the group;

[0059] Step 9 determines whether the iteration stop condition is met. If so, execute step 10; otherwise, increase the number of loops by 1 and execute step 3;

[0060] Step 10 outputs the saved optimal path.

[0061] The beneficial effect of the present invention is that the mayfly algorithm is improved based on a dynamic adaptive parameter adjustment strategy, the global optimal solution can be obtained, the efficiency and stability of path planning solution are improved, the balance between local search capability and global search capability is improved through dynamic adaptive parameter adjustment, the execution efficiency of the algorithm is accelerated, and the optimization ability of the algorithm is improved.

[0062] The effect of the present invention can be further illustrated by the following simulation experiment:

[0063] In order to verify the correctness and rationality of this method, the algorithm was simulated in a 20×20 grid environment model using Python programming and compared with the basic mayfly algorithm. The parameters in the algorithm are set as follows: in the traditional mayfly algorithm, the size of the male and female populations is set to 10, the inertia weight g = 0.8, the positive attraction coefficient a1 = 1.0, a2 = 1.5, a3 = 1.5, the random walk coefficient fl = 0.5, the maximum speed Velmax = 20, the minimum speed Velmin = -20, the maximum number of iterations T = 100, and the dance coefficient dance = 0.5. In the algorithm of this paper, g, a1, and a2 are set as adaptive parameters, and the other parameters are the same as above.

[0064] The simulation results are as follows Figure 2 and Figure 3 As shown. Analysis of the simulation results shows that the optimal path length obtained by the traditional mayfly algorithm is 32.6427, and the minimum number of iterations is 45 generations, while the path length obtained by the dynamic adaptive parameter adjustment mayfly algorithm is 28.3583, and the minimum number of iterations is 11 generations. In addition, the number of corners of the optimal paths obtained by the two algorithms are 4 and 3 respectively. Therefore, compared with the traditional mayfly algorithm, the dynamic adaptive parameter adjustment mayfly algorithm can find a better path at a faster speed, with fewer corners, and the resulting path is closer to the actual path.

[0065] In order to further verify the stability of the improved algorithm proposed in the present invention, the improved ant colony algorithm recorded in the article “Intelligent Car Path Planning Based on Ant Colony Algorithm” published in the journal “Computer Simulation” in 2021 was selected. The method of the present invention was used for simulation in the more complex grid environment of 30×30 recorded in the article.

[0066] The simulation results are as follows Figure 4 and Figure 5 As shown in the figure, it can be seen that the improved mayfly algorithm in the literature needs 60 generations to converge to the optimal solution 42; while the adaptive parameter adjustment mayfly algorithm of the present invention only needs 18 generations to find the optimal solution 41.7812. In addition, the number of corners of the optimal paths obtained by the two algorithms are 8 and 3 respectively.

[0067] Through comparative simulation, it can be concluded that the use of dynamic adaptive parameter adjustment strategy to improve the traditional mayfly algorithm can effectively improve the solution-finding effect of the algorithm in the early stage, increase the convergence speed of the algorithm, and the optimization results and smoothness are better than those of the traditional mayfly algorithm and the improved algorithm in the literature.

[0068] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form; any technician familiar with the art can make many possible changes and modifications to the technical solution of the present invention by using the above disclosed methods and technical contents without departing from the scope of the technical solution of the present invention, or modify it into an equivalent embodiment of equivalent changes. Therefore, any simple modification, equivalent replacement, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of protection of the technical solution of the present invention.

Claims

1. A mobile robot path planning method based on the mayfly algorithm with dynamic adaptive parameter adjustment, comprising the following steps: Step 1: Initialize the environment, digitally encode the motion environment, and map the encoded 0-1 matrix into a grid graph model; Step 2: Initialize the fixed parameters of the algorithm: the position x, speed v, number n of male and female mayflies, positive attraction coefficient a3 of female mayflies, visual acuity coefficient β, random walk coefficient fl, maximum speed Velmax, minimum speed Velmin, number of iterations T, and dance coefficient dance; Step 3: Update the adaptive parameter values: positive attraction coefficients a1, a2, inertia weight g. The settings of the adaptive parameters are as shown in formulas (8)-(10) Step 4: Evaluate individual fitness, find the individual optimal pbest and the group optimal position gbest, and calculate the fitness according to formula (1); Step 5: Update the position and speed of male and female mayflies through formulas (2)-(5) to evaluate fitness; where r p ,r g is x i and the Cartesian distance between the local optimum and the global optimum, e is a random number between [-1,1], r mf is the Cartesian distance between males and females; Step 6: Sort the male and female mayflies according to their fitness from low to high, and mate them in order according to formulas (6) and (7); offspring1=L*male+(1-L)*female (6) offspring2=L*female+(1-L)*male (7) Where male is the male parent, female is the female parent, and L is a random value in [0,1]; Step 7: randomly divide the offspring into males and females; Step 8: Sort the male and female mayflies according to their fitness, retain the top n excellent individuals, and update the optimal positions of the individuals and the group; Step 9 determines whether the iteration stop condition is met. If so, execute step 10; otherwise, increase the number of loops by 1 and execute step 3; Step 10 outputs the saved optimal path.

Images