A robot path planning method based on improved botryllus schlosseri colony model

Abstract

The present application relates to the technical field of path planning, and especially relates to a robot path planning method based on an improved sea squirts colony model, which comprises the following steps: establishing a navigation point map model, constructing a mobile robot path cost function and a collision cost function with the shortest path as a target, constructing an improved sea squirts model, and obtaining optimal path points by using the improved sea squirts model. The present application establishes a model with the shortest path as a target function model, and the model is free of collision obstacles. The original algorithm is improved by means of chaotic initialization of initial sea squirts colony, increase of inertia weight, and non-uniform mutation, so as to balance the algorithm development and exploration efficiency. Finally, the improved sea squirts optimization algorithm is combined with cubic spline interpolation, so as to better complete the robot path planning problem.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to a robot path planning method based on an improved tunicate swarm model. Background Technology

[0002] The salps algorithm, proposed by Australian researchers in 2017, is a novel swarm intelligence optimization algorithm that simulates the group foraging behavior of marine salps. Its mechanism is simple and easy to understand, and it is convenient and easy to implement. It has become a hot research topic for many researchers both domestically and internationally. This algorithm is now widely used in practical problems.

[0003] However, the current tunicate swarm algorithm also has drawbacks such as inaccurate searching in the later stages of iteration and poor population diversity, which limit the algorithm's local development capabilities and overall exploration capabilities. Compared with other intelligent optimization algorithms, in order to further improve the solution accuracy and efficiency of this algorithm, it is necessary to improve the optimization solution strategy of the tunicate swarm algorithm. Summary of the Invention

[0004] To address the shortcomings of existing methods, this invention establishes a model with the objective function of minimizing the generated path in a collision-free obstacle region. The original algorithm is improved by operations such as chaotic initialization of the initial tunicate swarm, adding inertial weights, and non-uniform mutation, achieving a balance between algorithm development and exploration efficiency. Finally, the improved tunicate optimization algorithm is combined with cubic spline interpolation to better solve the robot path planning problem.

[0005] The technical solution adopted in this invention is: a robot path planning method based on an improved tunicate swarm model, comprising the following steps:

[0006] Step 1: Establish a navigation point map model, and construct the path cost function and collision cost function for the mobile robot with the goal of finding the shortest path.

[0007] Furthermore, the navigation point map model defines the robot's movement scenario as a planar space containing several static obstacles, and sets the center and radius of each obstacle.

[0008] Furthermore, the formulas for the path cost function and the collision cost function are as follows:

[0009] F = L + D; L represents the path consumption cost, defined as follows:

[0010]

[0011] Among them, (x i ,y i (x) represents the coordinates of the current path point. i-1 ,y i-1() represents the coordinates of the previous path point; D represents the collision cost, defined as follows:

[0012]

[0013] In the formula, δ represents the weighting coefficient, which is usually taken as 100; r(d) represents the radius of the d-th obstacle; and d represents the number of obstacles.

[0014] Step 2: Construct an improved model of the tunicate;

[0015] Furthermore, specifically including:

[0016] Step 21: Initialize population location;

[0017] The initial position of the population is obtained by inverse mapping using the chaotic mapping PWLCM method; the formula for the chaotic mapping PWLCM method is:

[0018]

[0019]

[0020] Where ub and lb are the upper and lower boundaries of the search particle. Given a chaotic sequence in the interval [0,1], To search for particles in the newly generated tunicate;

[0021] Step 22: Improve leader position updates;

[0022] Furthermore, specifically including:

[0023] First, calculate the fitness value of each individual in the population, and then set the individual with the optimal fitness value as the food source location F. p The leader and followers comprise half of the population, and the leader guides the group toward the optimal value. The improved leader update formula is as follows:

[0024]

[0025] in, F represents the position of individual tunicate i in the exploration space after the t-th iteration. p This represents the location of the food source, and ω is the inertial weight.

[0026] The definition of ω is as follows:

[0027] ω=log(1+e -t (7)

[0028] Where t is the number of iterations.

[0029] Step 23: Improve follower location updates;

[0030] Furthermore, specifically including:

[0031]

[0032] in, This represents the position of the i-th follower in the j-dimensional space. Let J be the position in the j-dimensional space of the individual with the best fitness among the followers in iteration t-1. This represents the position of the individual in the j-dimensional space during the previous iteration.

[0033] Step 24: Introduce non-uniform variation to improve the model of tunicate colonies;

[0034] Furthermore, specifically including:

[0035] Let the population size be n and the spatial dimension be j, then The position of individual i in the j-dimensional space is represented by the non-uniform mutation operation applied to its nth component. The formula for generating a new individual is:

[0036]

[0037] In the formula, ub and lb are the upper and lower boundaries of the search particle; The nth component represents the position of individual i in the j-dimensional space; Where t represents the current iteration number; T is the maximum iteration number; r is a random number between [0, 1]; and b is the non-uniformity parameter.

[0038] Step 3: Use the improved tunicate model to obtain the optimal path point for the robot's movement.

[0039] Furthermore, specifically including:

[0040] The cubic spline interpolation method is used to interpolate between adjacent nodes. By connecting the starting point, all q nodes, n interpolation points, and the ending point, an ordered line segment is obtained.

[0041] The beneficial effects of this invention are:

[0042] 1. First, the population is initialized with chaos to increase its diversity; inertia weights are added to improve the leader update method, and the best individual is replaced with the previous generation individual, which improves the follower's process optimization ability; then, the newly generated individuals are subjected to non-uniform mutation operation, which effectively balances the model's early extensive exploration and late fine mining capabilities, improving solution performance and accuracy; finally, interpolation is used to smooth the generated path points and optimize an optimal path from the starting point to the target point; the method of this invention is simple, fast, and effectively solves the problems of slow convergence speed and easy getting trapped in local optima in current mobile robot path planning optimization algorithms. Attached Figure Description

[0043] Figure 1 This is a flowchart of the robot path planning method based on the improved tunicate swarm model of the present invention;

[0044] Figure 2 This is the optimal path diagram for robot planning in this invention. Detailed Implementation

[0045] The present invention will be further described below with reference to the accompanying drawings and embodiments. The drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0046] like Figure 1 As shown, a robot path planning method based on an improved tunicate swarm model includes the following steps:

[0047] Step 1: Establish a navigation point map model and construct a mobile robot objective function model based on the shortest path, including the path cost function and the collision cost function.

[0048] Navigation point map model: The robot's movement scenario is envisioned as a planar space with several static obstacles, the position of each obstacle is known, and they are represented by circumcircles.

[0049] The expression for the obstacle is:

[0050] (xa) 2 +(yb) 2 =r 2 (1)

[0051] In the formula, (a, b) represents the center of the obstacle, and r represents the radius of the obstacle.

[0052] The path planning cost function is: F = L + D; L represents the path consumption cost, defined as follows:

[0053]

[0054] Among them, (xi ,y i (x) represents the coordinates of the current path point. i-1 ,y i-1 () represents the coordinates of the previous path point.

[0055] D represents the collision cost, defined as follows:

[0056]

[0057] In the formula, δ represents the weighting coefficient, which is usually taken as 100; r(d) represents the radius of the d-th obstacle; and d represents the number of obstacles.

[0058] Step 2: Construct an improved model of the tunicate;

[0059] The parameters of the salver model are initialized, including population size, maximum number of iterations, and population dimension information, and a random initial population is generated according to the initialization conditions; each individual of the salver population represents a group, and the fitness value of the individual is calculated by substituting it into the cost function.

[0060] Step 21: Initialize population location:

[0061] The chaotic mapping PWLCM method is used to improve the uniformity and iteration speed of the tunicate model and to increase the coverage mapping space of the initial solution. The calculation formula is as follows:

[0062]

[0063]

[0064] Where ub and lb are the upper and lower boundaries of the search particle. Given a chaotic sequence in the interval [0,1], To search for particles in the newly generated tunicate;

[0065] The initial position of the population is obtained by inverse mapping according to equation (4). This chaotic mapping method can greatly increase the coverage of the initial solution space, allowing the population to approach the optimal solution more quickly, thereby accelerating the convergence speed of the algorithm.

[0066] Step 22: Leader position update;

[0067] First, calculate the fitness value of each individual in the population, and then set the individual with the optimal fitness value as the food source location F. p The leader and followers comprise half of the population, and the leader guides the group toward the optimal value. The improved leader update formula is as follows:

[0068]

[0069] in, F represents the position of individual tunicate i in the exploration space after the t-th iteration. p This represents the location of the food source, and ω is the inertial weight that decreases non-linearly with the number of iterations. By gradually reducing the weight parameter ω, the global search capability can be improved. This method can achieve the best results in the early and late stages of iteration and can improve the population search diversity of the algorithm.

[0070] The definition of ω is as follows:

[0071] ω=log(1+e -t (7)

[0072] Where t is the number of iterations.

[0073] Step 23: Improve follower location updates;

[0074]

[0075] in, This represents the position of the i-th follower in the j-dimensional space. Let J be the position in the j-dimensional space of the individual with the best fitness among the followers in iteration t-1. This represents the position of the individual in the j-dimensional space during the previous iteration.

[0076] This improvement guides the followers toward the global optimum. Moving the algorithm effectively avoids the decline in optimization efficiency that occurs when blindly following previous individuals in the original optimization algorithm.

[0077] Step 24: Introduce non-uniform variation to improve the method;

[0078] By employing a non-uniform mutation method to adaptively adjust the step size of the search range in the later stages of the search, the *Trichinella spiralis* model can be helped to escape local optima throughout the search process and overcome premature convergence. Let the population size be *n* and the spatial dimension be *j*, then... The position of individual i in the j-dimensional space is represented by the non-uniform mutation operation applied to its nth component. The formula for generating a new individual is:

[0079]

[0080] In the formula, ub and lb are the upper and lower boundaries of the search particle; The nth component represents the position of individual i in the j-dimensional space; Where t represents the current iteration number; T is the maximum iteration number; r is a random number between [0, 1]; b is the non-uniformity parameter; (t, y) enables the tunicate model to perform a large-scale search in the early stage of optimization and has the ability to escape local optima through the step size of non-uniform variation. During the iteration process, the search radius is reduced probabilistically so that the search around the current solution is refined in the later stage to obtain the global optimum as much as possible.

[0081] Return the optimal fitness value; determine if the termination condition is met. If yes, end the algorithm and output the result; otherwise, return to the previous step and continue iterating. The algorithm terminates when it reaches the maximum number of iterations.

[0082] The fitness value of each individual is calculated based on the cost function. The individual positions and food positions of the population are updated, and the next iteration begins. This process continues until the upper limit of iteration is reached, at which point the optimal individual position is obtained, which is the optimal path point.

[0083] Step 3: Obtain the optimal path point using the improved tunicate model.

[0084] Mapping the path to an individual in the improved tunicate model, each individual is represented as a j-dimensional array, where each dimension corresponds to the coordinates of a node on the path, i.e., composed of the coordinate values ​​of a path node. Thus, each individual represents a sequence of length d, the number of path nodes; assuming there are q nodes on the path, with coordinates (x1, y1), (x2, y2), ..., (x...). q ,y q The starting and ending coordinates of the path are (x0, y0) and (x0, y0) respectively. q+1 ,y q+1 Using cubic spline interpolation, interpolation is performed between adjacent nodes, generating a total of n interpolation points, whose coordinates can be represented as (x... i ,y i ), where i ranges from 1 to n; therefore, by connecting the starting point, all q nodes, n interpolation points, and the ending point, we obtain an ordered line segment, namely (x0, y0), (x1, y1), ..., (x i ,y i This indicates that the robot actually moves along the line segment during operation.

[0085] Example 1

[0086] Experimental verification was conducted, setting a 6*6 simulation space. The robot's starting and ending coordinates were (0,0) and (6,6) respectively. The number of obstacles in the scene space was set to 6, with coordinates (1.5,4.5), (4.0,4.0), (1.2,1.5), (3.1,2.6), (5.1,4.7), and (4.5,1.7), and threat radii of 1.0, 0.4, 0.8, 0.3, 0.6, and 0.5 respectively. The number of path nodes was set to 4, and the number of interpolation points was set to 300. The initial population size of the tunicate was set to 300, and the maximum number of iterations was set to 100. The results were obtained using Matlab as follows: Figure 2 The path shown.

[0087] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A robot path planning method based on an improved tunicate swarm model, characterized in that, Includes the following steps: Step 1: Establish a navigation point map model, and construct the path cost function and collision cost function for the mobile robot with the goal of finding the shortest path. Step 2: Construct an improved model of the tunicate swarm; Step two specifically includes: Step 21: Initialize population location; The initial position of the population is obtained by inverse mapping using the chaotic mapping PWLCM method; the formula for the chaotic mapping PWLCM method is: ； （ ）+ ； in, and It searches for the upper and lower boundaries of the particles. Given a chaotic sequence in the interval [0,1], To search for particles in the newly generated tunicate; Step 22: Improve leader position updates; Step 23: Improve follower location updates; Step 24: Introduce non-uniform variation to improve the model of tunicate colonies; Step 24 specifically includes: Let the population size be n If the spatial dimension is j, then Representative of individuals At its position in the j-dimensional space, for the first... Each component undergoes a non-uniform mutation operation, and the formula for generating a new individual is: ； In the formula, and It refers to the upper and lower boundaries of the search particle; Represents an individual The position of the j-th dimension in space One component; ;in, Indicates the current iteration number; This represents the maximum number of iterations. A random number between [0, 1] It is a non-uniformity parameter; Step 3: Use the improved tunicate group model to obtain the optimal path point for the robot's movement.

2. The robot path planning method based on the improved tunicate swarm model according to claim 1, characterized in that, The navigation point map model is a planar space where the robot's movement scenario is defined, containing several static obstacles, with the center and radius of each obstacle set.

3. The robot path planning method based on the improved tunicate swarm model according to claim 1, characterized in that, The formulas for the path cost function and the collision cost function are as follows: ; The cost of the path is represented as follows: ； in,( , () represents the coordinates of the current path point. , () represents the coordinates of the previous path point; Represented as collision cost, defined as follows: ； In the formula, δ represents the weighting coefficient, which is set to 100; Indicates the first The radius of the obstacle; Indicates the number of obstacles; Indicates the center of the circle representing the obstacle.

4. The robot path planning method based on the improved tunicate swarm model according to claim 1, characterized in that, Step 22 specifically includes: First, calculate the fitness value of each individual in the population, and then set the individual with the optimal fitness value as the food source location. The leader and followers comprise half of the population at the front and half at the back. The leader guides the group toward the optimal value. The improved leader position update formula is: (t)= ； in, (t) represents the i-th salps in the exploration space. Position after the next iteration This indicates the location of the food source. Inertial weight; The definition is as follows: = ； in, t This represents the number of iterations.

5. The robot path planning method based on the improved tunicate swarm model according to claim 1, characterized in that, Step 23 specifically includes: t = ； in, (t) represents the position of the i-th follower in the j-dimensional space. For followers The position of the best-fit individual in the j-dimensional space during iteration. This represents the position of the individual in the j-dimensional space during the previous iteration.

6. The robot path planning method based on the improved tunicate swarm model according to claim 1, characterized in that, Step three specifically includes: The cubic spline interpolation method is used to interpolate between adjacent nodes. By connecting the starting point, all q nodes, n interpolation points, and the ending point, an ordered line segment is obtained.

Images