Metro obstacle avoidance robot path planning method based on badger optimization algorithm

Abstract

The invention relates to a metro obstacle avoidance robot path planning method based on a badger optimization algorithm, and the method comprises the following steps: 1, building a two-dimensional plane model of a metro environment, dividing the metro environment into a grid map, determining a starting position and a target position of a robot, and the position and the shape of an obstacle, converting a path planning problem into a mathematical model; step 2, based on the mathematical model in the step 1, setting parameters of an improved badger optimization algorithm, including population quantity, position, olfactory intensity and density factor, and setting functions of a badger digging stage and a honey collection stage; and step 3, providing a reverse learning strategy to update the path. According to the method, the priori knowledge and constraint conditions of the subway environment are combined, the path better meeting the actual requirement can be planned for the subway obstacle avoidance robot, and the safety and feasibility of the path are improved.

Description

Technical Field

[0001] This invention relates to the field of subway robot path planning technology, and in particular to a subway obstacle avoidance robot path planning method based on the honey badger optimization algorithm. Background Technology

[0002] With the continuous development of subway systems, subway obstacle avoidance robots are playing an increasingly important role in ensuring subway operation safety, equipment maintenance, and troubleshooting. However, the subway environment is complex and ever-changing, with various obstacles and limitations. How to plan a safe, efficient, and accurate path for subway obstacle avoidance robots is a key problem that needs to be solved.

[0003] The subway path planning problem involves using intelligent algorithms to find a safe path from the starting point to the destination. Currently, commonly used robot path planning algorithms include ant colony optimization (ACO) and particle swarm optimization (PSO). However, these algorithms have some shortcomings when dealing with the subway obstacle avoidance robot path planning problem. For example, ACO converges slowly and is prone to stagnation; PSO is prone to getting trapped in local optima, resulting in unsatisfactory path length and calculation time.

[0004] The existing technology has the following problems:

[0005] 1. Insufficient real-time performance

[0006] Data update lag: The system relies on untimely updates of real-time data, resulting in inaccurate path planning.

[0007] Slow computation speed: Complex algorithms are time-consuming to compute in large-scale networks, making it difficult to respond in a timely manner.

[0008] 2. High algorithm complexity

[0009] High computational resource consumption: Complex algorithms require a large amount of computational resources, increasing costs.

[0010] Poor scalability: It is difficult to adapt to the needs of large-scale network expansion.

[0011] 3. Data quality and coverage issues

[0012] Inaccurate data: Errors or missing data affect planning results.

[0013] Incomplete coverage: Insufficient data in some areas leads to incomplete planning.

[0014] Existing subway route planning technologies have shortcomings in terms of real-time performance, data quality, algorithm complexity, and environmental adaptability, and need further optimization. Summary of the Invention

[0015] This invention provides a path planning method for subway obstacle avoidance robots based on the honey badger optimization algorithm, aiming to solve the problems of slow convergence speed, easy getting trapped in local optima, and low path planning accuracy in existing subway obstacle avoidance robot path planning algorithms. By improving the traditional honey badger optimization algorithm, introducing adaptive inertia weights and dynamic learning factors, and combining prior knowledge and constraints of the subway environment, efficient and accurate path planning for subway obstacle avoidance robots is achieved.

[0016] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a path planning method for a subway obstacle avoidance robot based on the honey badger optimization algorithm, characterized in that the method includes the following steps:

[0017] Step 1: Establish a two-dimensional planar model of the subway environment, divide the subway environment into a grid map, determine the robot's starting position and target position, as well as the position and shape of obstacles, and transform the path planning problem into a mathematical model;

[0018] Step 2: Based on the subway route planning mathematical model defined in Step 1, set the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor, and set the functions for the honey badger digging and honey-collecting stages;

[0019] Step 3: Propose a reverse learning strategy to update the path, and output the optimal path planning result according to the designed fitness function. The subway obstacle avoidance robot moves according to the path planning result, from the starting point to the destination, without colliding with obstacles along the way.

[0020] One possible design, wherein step 1 includes the following steps:

[0021] Step 1.1: Create a grid map of the working environment of the subway obstacle avoidance robot according to its own size, model the subway environment, and divide the subway environment into grids of equal size;

[0022] Step 1.2: Set passable locations to white grids and impassable locations to black grids;

[0023] Step 1.3: Set the starting point and ending point according to the subway working environment, determine the robot's starting position and target position, as well as the position and shape of obstacles.

[0024] One possible design, where step 2 includes the following steps:

[0025] Step 2.1: Initialize the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor;

[0026] Step 2.2: In the mining process, the honey badger performs a local search based on its current position and adaptive inertia weights to find a better position. The adaptive inertia weights are dynamically adjusted according to the number of iterations to balance the algorithm's global search and local search capabilities.

[0027] Step 2.3: During honey gathering, the honey badger performs a global search based on outstanding individuals and a dynamic learning factor. The dynamic learning factor is also dynamically adjusted according to the number of iterations to improve the convergence speed and accuracy of the algorithm.

[0028] One possible design, in step 2.1:

[0029] After initialization and improvement, the honey badger population size N>0, and the location formula is defined as follows:

[0030]

[0031] Where f(α) is the search range factor, α is a random number between 0 and 4, and x i Let i be the position of the i-th honey badger. and Let $\mathbf{i}$ be the upper and lower bounds of the search range for the $i$-th honey badger, respectively.

[0032] The formula for defining the olfactory intensity of the honey badger is:

[0033]

[0034] S i =π(x) i -x i+1 ) 2 (5)

[0035] d i =(x p -x i ) 2 (6)

[0036] Where g(β) is the olfactory range factor, β is a random number between -5 and 5, and S is the source intensity or density intensity of the i-th honey badger. Let i be the position of the i-th honey badger. Let d be the position of the (i+1)th honey badger. i Let x represent the distance between the i-th honey badger and its prey. p Indicates the location of the prey;

[0037] The density factor is defined as:

[0038] D=γe -t / T (7)

[0039] Where γ is the density range factor, 1 < γ < 2, t is the current iteration number, and T is the maximum iteration number.

[0040] One possible design, in step 2.2:

[0041] In the mining behavior, the formula for updating the honey badger's position is set as follows:

[0042]

[0043] Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. Let w be the optimal position of the honey badger in the t-th iteration, and w be the adaptive factor. max and w min These are the maximum and minimum values ​​of the adaptive factor w, respectively, where 0 < w min <w max .

[0044] One possible design, in step 2.3:

[0045] During honey gathering and digging activities, the honey badger's location update formula is:

[0046]

[0047]

[0048] Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. To find the optimal position of the honey badger in the t-th iteration, Let w1 and w2 be the random position of the honey badger in the t-th iteration, and w1 and w2 be adaptive factors. 1，max and w 1，min These are the maximum and minimum values ​​of the adaptive factor w1, respectively, where 0 < w 1，min <w 1，max w 2，max and w 2，min These are the maximum and minimum values ​​of the adaptive factor w2, respectively, where 0 < w 2，min <w 2，max .

[0049] One possible design, step 3 includes:

[0050] Step 3.1: Randomly generate an initial population, where each individual represents a possible robot path, and the path consists of a series of grid coordinates;

[0051] Step 3.2: Propose a reverse learning strategy to update the path;

[0052] Step 3.3: Sort the individuals according to their fitness values, select the individuals with better fitness values ​​as the best individuals, and output the optimal path planning result.

[0053] One possible design, in step 3.2:

[0054] After all honey badgers have updated their positions, to prevent the algorithm from getting stuck in local optima and failing to calculate a path, a back-learning strategy is set as follows:

[0055]

[0056] Where t is the current iteration number, Let a be the position of the i-th honey badger in the t-th iteration, and b be the learning range. k represents adaptive learning, where 0 < k < 1.

[0057] One possible design, in step 3.3:

[0058] To select the optimal path through iteration, the evaluation metric is set as a function based on path length and turning angle:

[0059]

[0060] Among them, t i The time consumed in the current iteration. Let i be the position of the i-th path point. Let be the position of the i-th path point.

[0061] The beneficial effects of this invention compared with the prior art are: by introducing adaptive inertia weights and dynamic learning factors, the search capability of the honey badger optimization algorithm is improved, the convergence speed and accuracy of the algorithm are increased, and the algorithm is prevented from getting trapped in local optima.

[0062] By combining prior knowledge and constraints of the subway environment, a path that better meets actual needs can be planned for the subway obstacle avoidance robot, improving the safety and feasibility of the path.

[0063] This invention can effectively solve the path planning problem of subway obstacle avoidance robots, improve the safety and reliability of subway operation, and has high practical value and application prospects. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 A schematic diagram of a raster map model for the subway environment;

[0066] Figure 2 A schematic diagram illustrating the position update process for improving the honey badger optimization algorithm. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] This invention proposes a path planning method for a subway obstacle avoidance robot based on the honey badger optimization algorithm. The method includes the following steps:

[0070] Step 1: Establish a two-dimensional planar model of the subway environment, divide the subway environment into a grid map, determine the robot's starting position and target position, as well as the position and shape of obstacles, and transform the path planning problem into a mathematical model;

[0071] Step 2: Based on the subway path planning mathematical model defined in Step 1, design a subway obstacle avoidance robot path planning method based on the honey badger optimization algorithm, set the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor, and set the functions for the honey badger digging stage and honey collection stage.

[0072] Step 3: Propose a reverse learning strategy to update the path, and output the optimal path planning result according to the designed fitness function. The subway obstacle avoidance robot moves according to the path planning result, from the starting point to the destination, without colliding with obstacles along the way.

[0073] Furthermore, step 1 specifically includes the following steps:

[0074] Step 1.1: Create a grid map of the subway obstacle avoidance robot's working environment based on its own size, and model the subway environment. For example... Figure 2 As shown, the subway environment is divided into grids of equal size;

[0075] Step 1.2: Passable locations are set to white grids, and impassable locations are set to black grids. In the program, 0 represents passable areas and 1 represents obstacle areas (impassable locations).

[0076] Step 1.3: Set the starting point and ending point according to the subway working environment, determine the robot's starting position S and target position T, as well as the position and shape of obstacles.

[0077] Furthermore, step 2 specifically includes the following steps:

[0078] Step 2.1: Initialize the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor.

[0079] Furthermore, after initialization and improvement, the honey badger population size N>0, and the location formula is defined as follows:

[0080]

[0081] Where f(α) is the search range factor, α is a random number between 0 and 4, and x i Let i be the position of the i-th honey badger. and Let $\mathbf{i}$ be the upper and lower bounds of the search range for the $i$-th honey badger, respectively.

[0082] Furthermore, the formula for defining the honey badger's olfactory intensity is as follows:

[0083]

[0084] S i =π(x i -x i+1 ) 2 (5)

[0085] d i =(x p -x i ) 2 (6)

[0086] Where g(β) is the olfactory range factor, β is a random number between -5 and 5, and S is the source intensity or density intensity of the i-th honey badger. Let i be the position of the i-th honey badger. Let d be the position of the (i+1)th honey badger. i Let x represent the distance between the i-th honey badger and its prey.p Indicates the location of the prey.

[0087] Furthermore, the density factor is defined as

[0088] D=γe -t / T (7)

[0089] Where γ is the density range factor, 1 < γ < 2, t is the current iteration number, and T is the maximum iteration number.

[0090] Step 2.2: During the mining process, the honey badger performs a local search based on its current position and adaptive inertia weights to find a better location. The adaptive inertia weights are dynamically adjusted based on the number of iterations to balance the algorithm's global and local search capabilities.

[0091] In further settings for the mining behavior, the honey badger's position update formula is as follows:

[0092]

[0093] Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. Let w be the optimal position of the honey badger in the t-th iteration, and w be the adaptive factor. max and w min These are the maximum and minimum values ​​of the adaptive factor w, respectively, where 0 < w min <w max .

[0094] Step 2.3: During honey gathering, the honey badger performs a global search based on outstanding individuals and a dynamic learning factor. The dynamic learning factor is also dynamically adjusted according to the number of iterations to improve the convergence speed and accuracy of the algorithm.

[0095] In further honey-gathering and digging activities, the honey badger's location update formula is:

[0096]

[0097] Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. To find the optimal position of the honey badger in the t-th iteration, Let w1 and w2 be the random position of the honey badger in the t-th iteration, and w1 and w2 be adaptive factors. 1，max and w 1，min These are the maximum and minimum values ​​of the adaptive factor w1, respectively, where 0 < w 1，min <w 1，max w 2，max and w 2，minThese are the maximum and minimum values ​​of the adaptive factor w2, respectively, where 0 < w 2，min <w 2，max .

[0098] Furthermore, step 3 specifically includes the following steps:

[0099] Step 3.1: Randomly generate an initial population, where each individual represents a possible robot path. Each path consists of a series of grid coordinates.

[0100] Step 3.2: Propose a reverse learning strategy to update the path.

[0101] Furthermore, after all honey badgers have updated their positions, to prevent the algorithm from getting stuck in local optima and failing to calculate a path, a back-learning strategy is set as follows:

[0102]

[0103] Where t is the current iteration number, Let a be the position of the i-th honey badger in the t-th iteration, and b be the learning range. k represents adaptive learning, where 0 < k < 1.

[0104] Step 3.3: Sort the individuals according to their fitness values, select the individuals with better fitness values ​​as the best individuals, update the honey badger's position, and output the optimal path planning result.

[0105] Furthermore, in order to select the optimal path through iteration, the evaluation metric is set as a function based on path length and turning angle:

[0106]

[0107] Among them, t i The time consumed in the current iteration. Let i be the position of the i-th path point. Let be the position of the i-th path point.

[0108] like Figure 2 As shown, to illustrate the path planning performance of a subway obstacle avoidance robot based on an improved honey badger optimization algorithm, based on... Figure 1 The subway environment map is set with a starting point S(1,11) and an ending point T(20,11), where the honey badger population size N=500. γ = 1.5, maximum number of iterations T = 1000, w max =1.5, w min =1, w 1，max =2, w 1，min =1, w 2，max =3, w2，min =2, a=0, b=10 and k=0.9.

Claims

1. A path planning method for a subway obstacle avoidance robot based on the honey badger optimization algorithm, characterized in that, The method includes the following steps: Step 1: Establish a two-dimensional planar model of the subway environment, divide the subway environment into a grid map, determine the robot's starting position and target position, as well as the position and shape of obstacles, and transform the path planning problem into a mathematical model; Step 2: Based on the subway route planning mathematical model defined in Step 1, set the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor, and set the functions for the honey badger digging and honey-gathering stages; Step 3: Propose a reverse learning strategy to update the path, and output the optimal path planning result according to the designed fitness function. The subway obstacle avoidance robot moves according to the path planning result, from the starting point to the destination, without colliding with obstacles along the way.

2. The path planning method according to claim 1, characterized in that, Step 1 includes the following steps: Step 1.1: Create a grid map of the working environment of the subway obstacle avoidance robot according to its own size, model the subway environment, and divide the subway environment into grids of equal size; Step 1.2: Set passable locations to white grids and impassable locations to black grids; Step 1.3: Set the starting point and ending point according to the subway working environment, determine the robot's starting position and target position, as well as the position and shape of obstacles.

3. The path planning method according to claim 1, characterized in that, Step 2 includes the following steps: Step 2.1: Initialize the parameters of the improved honey badger optimization algorithm, including population size, location, olfactory intensity, and density factor; Step 2.2: In the mining process, the honey badger performs a local search based on its current position and adaptive inertia weights to find a better position. The adaptive inertia weights are dynamically adjusted according to the number of iterations to balance the algorithm's global search and local search capabilities. Step 2.3: During honey gathering, the honey badger performs a global search based on outstanding individuals and a dynamic learning factor. The dynamic learning factor is also dynamically adjusted according to the number of iterations to improve the convergence speed and accuracy of the algorithm.

4. The path planning method according to claim 3, characterized in that, In step 2.1: After initialization and improvement, the honey badger population size N>0, and the location formula is defined as follows: Where f(α) is the search range factor, α is a random number between 0 and 4, and x i Let i be the position of the i-th honey badger. and Let $\mathbf{i}$ be the upper and lower bounds of the search range for the $i$-th honey badger, respectively. The formula for defining the olfactory intensity of the honey badger is: S i =π(x i -x i+1 ) 2 (5) d i =(x p -x i ) 2 (6) Where g(β) is the olfactory range factor, β is a random number between -5 and 5, and S is the source intensity or density intensity of the i-th honey badger. Let i be the position of the i-th honey badger. Let d be the position of the (i+1)th honey badger. i Let x represent the distance between the i-th honey badger and its prey. p Indicates the location of the prey; The density factor is defined as: D=γe -t / T (7) Where γ is the density range factor, 1 < γ < 2, t is the current iteration number, and T is the maximum iteration number.

5. The path planning method according to claim 3 or 4, characterized in that, In step 2.2: In the mining behavior, the formula for updating the honey badger's position is set as follows: Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. Let w be the optimal position of the honey badger in the t-th iteration, and w be the adaptive factor. max and w min These are the maximum and minimum values ​​of the adaptive factor w, respectively, where 0 < w min <w max .

6. The path planning method according to claim 3 or 4, characterized in that, In step 2.3: During honey gathering and digging activities, the honey badger's location update formula is: Where t is the current iteration number and T is the maximum iteration number. Let i be the position of the i-th honey badger in the t-th iteration. To find the optimal position of the honey badger in the t-th iteration, Let w1 and w2 be the random position of the honey badger in the t-th iteration, and w1 and w2 be adaptive factors. 1，max and w 1，min These are the maximum and minimum values ​​of the adaptive factor w1, respectively, where 0 < w 1，min <w 1，max w 2，max and w 2，min These are the maximum and minimum values ​​of the adaptive factor w2, respectively, where 0 < w 2，min <w 2，max .

7. The path planning method according to claim 1, characterized in that, Step 3 includes: Step 3.1: Randomly generate an initial population, where each individual represents a possible robot path, and the path consists of a series of grid coordinates; Step 3.2: Propose a reverse learning strategy to update the path; Step 3.3: Sort the individuals according to their fitness values, select the individuals with better fitness values ​​as excellent individuals, and output the optimal path planning result.

8. The path planning method according to claim 7, characterized in that, In step 3.2: After all honey badgers have updated their positions, to prevent the algorithm from getting stuck in local optima and failing to calculate a path, a back-learning strategy is set as follows: Where t is the current iteration number, Let a be the position of the i-th honey badger in the t-th iteration, and b be the learning range. k represents adaptive learning, where 0 < k < 1.

9. The path planning method according to claim 7 or 8, characterized in that, In step 3.3: To select the optimal path through iteration, the evaluation metric is set as a function based on path length and turning angle: Among them, t i The time consumed in the current iteration. Let i be the position of the i-th path point. Let be the position of the i-th path point.

Images