Robot path planning method and system based on artificial bee colony and reinforcement learning

Abstract

The application provides a robot path planning method and system based on artificial bee colony and reinforcement learning, and relates to the field of path planning, and the method specifically comprises the following steps: initializing an artificial bee colony algorithm population according to known environment information, a starting point and a target point; iteratively optimizing the population based on a search mechanism of the artificial bee colony algorithm; when a preset optimization process termination condition is met, selecting a batch of high-quality solutions from the final population to form an elite solution set; encoding the elite solution set as prior knowledge of a reinforcement learning agent, initializing a value function of the reinforcement learning, and setting a reward function that combines path length and smoothness objectives; based on the initialized value function, running a reinforcement learning algorithm for policy search and optimization to fine-tune the path strategy; and outputting an optimal path after the reinforcement learning algorithm converges. The technical scheme of the application overcomes the problem in the prior art that a short and smooth path cannot be quickly generated in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of path planning, and more specifically to a robot path planning method and system based on artificial bee colony and reinforcement learning. Background Technology

[0002] Existing robot path planning methods can be mainly divided into traditional algorithms, heuristic algorithms, and emerging machine learning methods. Traditional graph search algorithms, such as A* and Dijkstra's algorithms, can guarantee finding the shortest path, but their computational complexity increases exponentially with environmental complexity and map resolution, resulting in poor real-time performance and difficulty in directly optimizing complex metrics such as path smoothness. Heuristic intelligent optimization algorithms, such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and the Artificial Bee Colony (ABC) algorithm involved in this invention, search for the global optimum by simulating natural phenomena, demonstrating good global search capabilities and flexibility in solving complex path planning problems. However, these algorithms also have inherent drawbacks: their convergence speed is often slow, especially when the solution space has high dimensionality and complex constraints, requiring a large number of iterative calculations to converge; in addition, the algorithm tends to linger near local optima in the later stages, leading to a decrease in population diversity and reduced optimization efficiency.

[0003] In recent years, reinforcement learning (RL) has attracted widespread attention in the field of path planning due to its powerful sequential decision-making capabilities. By interacting with the environment and being guided by reward functions, RL agents can learn efficient path policies. However, deep reinforcement learning typically requires a large amount of sampling and data training, resulting in a slow learning process; while tabular methods such as Q-learning face the "curse of dimensionality" problem when the state space is large. More importantly, the performance of RL algorithms is highly dependent on the initial policy and exploration efficiency. In the early stages of completely random exploration, its learning efficiency is low, and it is difficult to guarantee convergence to the global optimum.

[0004] Therefore, a prominent problem in existing technologies is that while single global heuristic algorithms (such as ABC) can conduct extensive exploration, they suffer from slow convergence in later stages and insufficient local development capabilities; while single reinforcement learning methods excel at fine-grained local decision-making, their initial exploration is blind, their global search efficiency is low, and they struggle to simultaneously optimize multiple objectives (such as shortest path and fewest inflection points). How to combine the advantages of both types of algorithms to design a hybrid optimization method that can quickly and efficiently plan both short and smooth paths has become a pressing technical challenge in this field.

[0005] Therefore, there is a need for a robot path planning method and system based on artificial bee colony and reinforcement learning that can guarantee the quality and efficiency of path planning and quickly generate short and smooth optimized paths in complex environments. Summary of the Invention

[0006] The main objective of this invention is to provide a robot path planning method and system based on artificial bee colony and reinforcement learning, so as to solve the problem that existing technologies cannot quickly generate short and smooth paths in complex environments.

[0007] To achieve the above objectives, this invention provides a robot path planning method based on artificial bee colony and reinforcement learning, specifically including the following steps:

[0008] S1. Initialize the artificial bee colony algorithm population based on the known environmental information, starting point, and target point.

[0009] S2 uses an artificial bee colony algorithm-based search mechanism to iteratively optimize the population.

[0010] S3. When the preset optimization process termination condition is met, a batch of high-quality solutions are selected from the final population to form an elite solution set.

[0011] S4 encodes the elite solution set into prior knowledge for the reinforcement learning agent, which is used to initialize the value function of reinforcement learning and to set the state space, action space, and reward function that combines path length and smoothness objectives.

[0012] S5, based on the initialized value function, runs a reinforcement learning algorithm to search for and optimize policies, so as to fine-tune the path policy.

[0013] S6 outputs the optimal path after the reinforcement learning algorithm converges.

[0014] Furthermore, step S1 specifically includes the following steps:

[0015] S1.1, each feasible path It consists of a series of node sequences, represented as:

[0016] ;

[0017] in, Starting point For the target point, For nodes in the environment.

[0018] S1.2, use the artificial bee colony algorithm for global search; initialize the ABC population with a population size of... Each nectar source represents a path, containing The number of nodes; the encoding method for honey sources uses integer encoding, the first node... The initialization formula for a honey source is:

[0019] ;

[0020] in, nectar source The first in One node; The generation rules are as follows:

[0021] ;

[0022] in, It is the set of all nodes in the environment, excluding the start and end points; , This represents the total number of nodes. This means randomly selecting a node from the set of nodes.

[0023] Furthermore, step S2 specifically includes the following steps:

[0024] S2.1, using the fitness function to comprehensively consider path length and smoothness, the fitness function... The expression is:

[0025] ;

[0026] in, The total length of the path is expressed by the following formula:

[0027] ;

[0028] in, Indicates Euclidean distance; Representing a path The number of inflection points; This is the smoothness penalty coefficient.

[0029] S2.2, Smoothness is reflected by the number of inflection points. The number of inflection points is determined by calculating the included angle between the vectors of three consecutive points. Let three consecutive points... , , This forms two vectors. and The cosine of the angle between two vectors for:

[0030] ;

[0031] like ,but This is determined to be an inflection point; among them, For the preset threshold, The larger the value, the weaker the strength of the inflection point determination.

[0032] S2.3, based on fitness function Neighborhood search and population update are performed; the formula for generating new nectar sources during the hired bee stage is:

[0033] ;

[0034] in, For new honey sources; This is the current honey source; The honey source is randomly selected from the area; It is a random number in the range [-1, 1].

[0035] Furthermore, step S3 specifically includes:

[0036] When the number of iterations of the ABC algorithm reaches the preset maximum value Or the optimal fitness of the population in a continuous Intra-generation improvement less than threshold When the time is right, terminate the ABC algorithm search; select the population with the highest fitness from the final population. A set of non-repeating paths constitutes the elite solution set. The screening formula is:

[0037] ;

[0038] in, For fitness threshold, For the first One path.

[0039] Furthermore, step S4 specifically includes the following steps:

[0040] S4.1, for each path in the elite solution set traverse each path State-Action Pairs ,in, Current node state, action To select the next node; the state-action pair The value is initialized to a priority value. :

[0041] ;

[0042] in, This is the action value function.

[0043] S4.2, Setting up the reinforcement learning environment, state space The set of all nodes; action space From state The set of reachable neighboring nodes; the reward function is a fusion of path length and smoothness objectives, reward function Designed as follows:

[0044] ;

[0045] in, The distance traveled; For indicator functions, if in If an inflection point is found, return 1; otherwise, return 0. The inflection point penalty weight.

[0046] S4.3 uses the Q-learning algorithm for policy fine-tuning. The value update formula is:

[0047] ;

[0048] in, and These represent the current state and action, respectively. Represents the execution of actions The next state after that; Indicates the next state Below, one of the action variables among all possible actions that the agent can perform; Represents the next state Of all possible actions The largest value; Indicates an immediate reward; The learning rate; This is the discount factor.

[0049] Furthermore, step S5 specifically includes the following steps:

[0050] S5.1, Q-learning employs an ε-greedy strategy to balance exploration and exploitation. The decision-making process of the ε-greedy strategy is described as follows:

[0051] ;

[0052] in, , representing the exploration rate; Indicates that the current The action variable with the largest value.

[0053] S5.2, when The change in the value function is less than the convergence threshold. ,Right now:

[0054] ;

[0055] Or reach the maximum number of training rounds The learning process will terminate at that time. For the number of iterations, For the first After the next iteration Value function.

[0056] Furthermore, step S6 specifically includes:

[0057] From the starting point Initially, the agent chooses the option with the greatest [potential] at each step. Value action :

[0058] ;

[0059] Until the target point is reached Generate the final path list .in, The first in the optimal path list Each node.

[0060] This invention also provides a robot path planning system based on artificial bee colony and reinforcement learning, comprising:

[0061] The population initialization module is used to initialize the artificial bee colony algorithm population based on known environmental information, starting point and target point. Each individual in the population represents a feasible path from the starting point to the target point, which is composed of path points.

[0062] The global search optimization module is used to iteratively optimize the population based on the search mechanism of the artificial bee colony algorithm. The quality of individuals is evaluated by a fitness function that comprehensively considers path length and path smoothness, and the population update is guided accordingly.

[0063] The elite solution set construction module is used to select a batch of high-quality solutions from the final population to form an elite solution set when the global search optimization module meets the preset optimization process termination conditions.

[0064] The reinforcement learning initialization module integrates reinforcement learning algorithms to encode the elite solution set into prior knowledge of the reinforcement learning agent, which is used to initialize the value function and set the state space, action space, and reward function that combines path length and smoothness objectives.

[0065] The policy fine-tuning module is used to run reinforcement learning algorithms to search for and optimize policies based on the initialized value function, so as to fine-tune the path policy.

[0066] The path output module is used to output the optimal path after convergence. The optimal path satisfies the requirements of shortest length and fewest inflection points.

[0067] The present invention has the following beneficial effects:

[0068] 1. This invention achieves an organic combination of global exploration and local development through a hierarchical optimization architecture, which not only leverages the global search capability of the artificial bee colony algorithm but also utilizes the local optimization advantages of the reinforcement learning algorithm, effectively avoiding the local optimum problem.

[0069] 2. This invention designs a multi-objective fitness function and reward function that integrates path length and smoothness, which can significantly reduce the number of inflection points while ensuring the shortest path, generating a smoother path that better conforms to the robot's motion characteristics, thereby improving motion efficiency and service life.

[0070] 3. This invention proposes an elite solution set screening and transformation mechanism, which transforms the high-quality solutions obtained by the artificial bee colony algorithm into prior knowledge for the reinforcement learning algorithm, greatly improving the initial performance and learning efficiency of the reinforcement learning algorithm and accelerating the overall convergence speed.

[0071] 4. This invention solves the problem of slow convergence speed or low solution quality of single algorithms in complex path planning. It is suitable for industrial robots equipped with multi-line laser scanners to perform automated scanning operations on the surface of complex workpieces, and also has application value in scenarios such as logistics distribution and intelligent inspection. Attached Figure Description

[0072] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0073] Figure 1 A flowchart of a robot path planning method based on artificial bee colony and reinforcement learning according to the present invention is shown.

[0074] Figure 2 This illustration shows an application scenario of an industrial robot equipped with a multi-line laser scanner scanning a workpiece, as described in an embodiment of the present invention.

[0075] Figure 3 This is a schematic diagram illustrating the generation of sampling points on the surface of the workpiece under test in a scanning and measurement robot environment as described in this embodiment of the invention.

[0076] Figure 4This is a schematic diagram of the planned scanning viewpoint set on the surface of the workpiece under test in a scanning measurement robot environment involved in the embodiments of the present invention.

[0077] Figure 5 This is a schematic diagram of the optimal path generated by the robot path planning method based on artificial bee colony and reinforcement learning in an embodiment of the present invention.

[0078] Figure 6 This is a schematic diagram showing the final scanning effect of an industrial robot equipped with a multi-line laser scanner on the surface morphology of a workpiece, generated according to the robot path planning method based on artificial bee colony and reinforcement learning as described in an embodiment of the present invention.

[0079] The reference numerals in the above figures are explained as follows:

[0080] 1. Industrial robot; 2. Multi-line laser scanner; 3. Binocular tracking camera; 4. Workpiece; 5. Computer. Detailed Implementation

[0081] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0082] Example 1

[0083] like Figure 1 The robot path planning method based on artificial bee colony and reinforcement learning shown includes the following steps:

[0084] S1. Initialize the artificial bee colony algorithm population based on the known environmental information, starting point, and target point.

[0085] S2 uses an artificial bee colony algorithm-based search mechanism to iteratively optimize the population.

[0086] S3. When the preset optimization process termination condition is met, a batch of high-quality solutions are selected from the final population to form an elite solution set.

[0087] S4 encodes the elite solution set into prior knowledge for the reinforcement learning agent, which is used to initialize the value function of reinforcement learning and to set the state space, action space, and reward function that combines path length and smoothness objectives.

[0088] S5, based on the initialized value function, runs a reinforcement learning algorithm to search for and optimize policies, so as to fine-tune the path policy.

[0089] S6 outputs the optimal path after the reinforcement learning algorithm converges.

[0090] Specifically, step S1 includes the following steps:

[0091] S1.1, each feasible path It consists of a series of node sequences, represented as:

[0092] ;

[0093] in, Starting point For the target point, For nodes in the environment.

[0094] S1.2, the Artificial Bee Colony (ABC) algorithm is used for global search; the ABC population is initialized with a population size of [missing value]. Each nectar source (i.e., individual nectar) represents a path, containing The number of nodes; the encoding method for honey sources uses integer encoding, the first node... The initialization formula for a honey source is:

[0095] ;

[0096] in, nectar source The first node in nectar source The second node in the sequence, and so on; nectar source The first in One node; The generation rules are as follows:

[0097] ;

[0098] in, It is the set of all nodes in the environment, excluding the start and end points; , This represents the total number of nodes. This means randomly selecting a node from the set of nodes.

[0099] The search mechanism based on the artificial bee colony algorithm iteratively optimizes the population, where the quality of individuals is evaluated by a fitness function that comprehensively considers path length and path smoothness, and this evaluation guides the population update.

[0100] Specifically, step S2 includes the following steps:

[0101] S2.1, the iterative process of the ABC algorithm uses a multi-objective fitness function to evaluate the quality of each honey source. The fitness function comprehensively considers path length and smoothness. The expression is:

[0102] ;

[0103] in, The total length of the path is expressed by the following formula:

[0104] ;

[0105] in, Indicates Euclidean distance; Representing a path The number of inflection points; This is the smoothness penalty coefficient, used to adjust the weight of path length and smoothness.

[0106] S2.2, Smoothness is reflected by the number of inflection points. The number of inflection points is determined by calculating the included angle between the vectors of three consecutive points. Let three consecutive points... , , This forms two vectors. and The cosine of the angle between two vectors for:

[0107] ;

[0108] like (in (This is a preset threshold, typically set to 30°~45°). This is determined to be an inflection point; among them, For the preset threshold, The larger the value, the weaker the strength of the inflection point determination.

[0109] S2.3, the ABC algorithm utilizes a cooperative search mechanism involving hired bees, observer bees, and scout bees, based on a fitness function. Neighborhood search and population update are performed; the formula for generating new nectar sources during the hired bee stage is:

[0110] ;

[0111] in, For new honey sources; This is the current honey source; The honey source is randomly selected from the area; It is a random number in the range [-1, 1].

[0112] Specifically, step S3 is as follows:

[0113] When the number of iterations of the ABC algorithm reaches the preset maximum value Or the optimal fitness of the population in a continuous Intra-generation improvement less than threshold hour, The result is a positive integer. Terminate the ABC algorithm search; select the population with the highest fitness from the final population. A set of non-repeating paths constitutes the elite solution set. The screening formula is:

[0114] ;

[0115] in, For fitness threshold, For the first One path.

[0116] Specifically, step S4 includes the following steps:

[0117] S4.1, Elite Solution Set The prior knowledge encoded into the reinforcement learning agent is used for initialization. Value. For each path in the elite solution set. traverse each path State-Action Pairs ,in, Current node state, action To select the next node; the state-action pair The value is initialized to a priority value. :

[0118] ;

[0119] in, This is the action value function.

[0120] S4.2, Setting up the reinforcement learning environment, state space The set of all nodes; action space From state The set of reachable neighboring nodes; the reward function is a fusion of path length and smoothness objectives, reward function Designed as follows:

[0121] ;

[0122] in, The distance traveled; For indicator functions, if in If an inflection point is found, return 1; otherwise, return 0. The inflection point penalty weight.

[0123] The reward function that integrates path length and smoothness objectives refers to the immediate reward value obtained by the agent after performing an action. Its value is negatively correlated with the path segment length increased by the current action and negatively correlated with whether the current action introduces an inflection point.

[0124] S4.3 uses the Q-learning algorithm for policy fine-tuning. The value update formula is:

[0125] ;

[0126] in, and These represent the current state and action, respectively. Represents the execution of actions The next state after that; Indicates the next state Below, one of the action variables among all possible actions that the agent can perform; Represents the next state Of all possible actions The largest value; Indicates an immediate reward; The learning rate; This is the discount factor.

[0127] Specifically, step S5 includes the following steps:

[0128] S5.1, Q-learning employs an ε-greedy strategy to balance exploration and exploitation. The decision-making process of the ε-greedy strategy is described as follows:

[0129] ;

[0130] in, , representing the exploration rate, indicates the probability that the agent will randomly select an action to explore; Indicates that the current The action variable with the largest value.

[0131] S5.2, when The change in the value function is less than the convergence threshold. ,Right now:

[0132] ;

[0133] Or reach the maximum number of training rounds The learning process will terminate at that time. For the number of iterations, For the first After the next iteration Value function.

[0134] Specifically, step S6 is as follows:

[0135] From the starting point Initially, the agent chooses the option with the greatest [potential] at each step. Value action :

[0136] ;

[0137] Until the target point is reached Generate the final path list .in, The first in the optimal path list Each node. Both of the following conditions must be met:

[0138] 1. Shortest length: 2. Fewest inflection points: .

[0139] To verify the method of this invention, path planning was performed during the scanning operation of a workpiece with freeform surface features on an industrial robot equipped with a multi-line laser scanner. For example... Figure 2 As shown, a six-axis industrial robot 1 has a multi-line laser scanner 2 mounted on its end flange. A workpiece 4, with complex free-form surface features, is placed directly below it. A binocular tracking camera 3 is located on the right, used to identify reflective points on the multi-line laser scanner to determine its position and orientation in three-dimensional space. A computer 5 communicates with both the multi-line laser scanner and the industrial robot. The industrial robot moves according to a robot path planning method based on a hybrid optimization mechanism of artificial bee colony-reinforcement learning, obtaining the optimal path point set. Simultaneously, the multi-line laser scanner transmits the scanned data to the computer in real time for display, ultimately verifying the effectiveness of the invention.

[0140] First, the surface features of the workpiece are identified and divided using a surface sampling method, resulting in a set of small surface patches. Then, the center points of each small surface patch are obtained, such as... Figure 3 As shown, the normal vector of the point with respect to the workpiece surface is calculated; finally, based on the optimal scanning distance of the laser scanner, the final set of scanning viewpoints, i.e., the set of path points to be planned, is obtained, as shown. Figure 4 As shown.

[0141] The method provided by this invention processes the set of points to be planned to obtain the final path, such as... Figure 5 As shown. According to Figure 5The generated scanning path ultimately yields the scanning results of the workpiece surface morphology, with significant effects and good reconstruction results, such as... Figure 6 As shown.

[0142] Example 2

[0143] A robot path planning system based on artificial bee colony and reinforcement learning includes:

[0144] The population initialization module is used to initialize the artificial bee colony algorithm population based on known environmental information, starting point, and target point. Each individual in the population represents a feasible path from the starting point to the target point, consisting of path points.

[0145] The global search optimization module is used to iteratively optimize the population based on the search mechanism of the artificial bee colony algorithm. The quality of individuals is evaluated by a fitness function that comprehensively considers path length and path smoothness, and the population update is guided accordingly.

[0146] The elite solution set construction module is used to select a batch of high-quality solutions from the final population to form an elite solution set when the global search optimization module meets the preset optimization process termination conditions.

[0147] The reinforcement learning initialization module integrates reinforcement learning algorithms to encode the elite solution set into prior knowledge for the reinforcement learning agent, which is used to initialize the value function and set the state space, action space, and reward function that combines path length and smoothness objectives.

[0148] The policy fine-tuning module is used to run reinforcement learning algorithms to search for and optimize policies based on the initialized value function, so as to fine-tune the path policy.

[0149] The path output module is used to output the optimal path after convergence. The optimal path satisfies the requirements of shortest length and fewest inflection points.

[0150] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A robot path planning method based on artificial bee colony and reinforcement learning, characterized in that, Specifically comprising the following steps: S1, initializing a population of artificial bee colony algorithm according to known environment information, a starting point and a target point; S2, iteratively optimizing the population based on a search mechanism of the artificial bee colony ABC algorithm; S3, when a preset optimization process termination condition is met, selecting a batch of high-quality solutions from the final population to form an elite solution set; S4, encoding the elite solution set as prior knowledge of a reinforcement learning agent to initialize a value function of the reinforcement learning, and setting a state space, an action space and a reward function that combines path length and smoothness objectives; S5, based on the initialized value function, running a reinforcement learning algorithm to search and optimize strategies to fine-tune the path strategy; S6, outputting an optimal path after the reinforcement learning algorithm converges; Step S4 specifically comprises the following steps: S4.1, for each path in the elite set , traverse each state-action pair of the path where, is the current node state, action is the next node to select; initialize the value of the state-action pair to a priority value :​ ； wherein, is the action value function; S4.2, Set up the environment of reinforcement learning, state space is the set of all nodes; action space is the set of reachable adjacent nodes from state is the reward function that combines path length and smoothness objectives is designed as: ； wherein, is the distance moved; is an indicator function that returns 1 if a kink is produced at and 0 otherwise; is a kink penalty weight; S4.3, fine-tuning the strategy by using Q-learning algorithm, The value update formula is: wherein, represents performing an action the next state after; represents the next state next, one of all possible actions that the agent can perform the action variable; represents the next state all possible actions the maximum value; represents the immediate reward; is the learning rate; is the discount factor; Step S5 specifically comprises the following steps: S5.1, Q-learning uses an ε-greedy strategy to balance exploration and exploitation, and the decision-making process of the ε-greedy strategy is expressed as: ； wherein, represents an exploration rate; denotes the action variable that maximizes the current value; S5.2, when the value function change is less than the convergence threshold i.e.: ； or a maximum number of training rounds is reached the learning process is terminated; wherein, is the number of iterations, is the value function after the th iteration, th iteration. Step S6 specifically is: From the starting point At the beginning, the agent selects at each step the action with the largest value. ； until the target point is reached , generating a final path list ; wherein, is the th node in the optimal path list.

2. The robot path planning method based on artificial bee colony and reinforcement learning according to claim 1, characterized in that, Step S1 specifically comprises the following steps: S1.1, each feasible path consists of a sequence of nodes, denoted as: ； wherein, is a starting point, is a target point, is a node in the environment; S1.2, global search is performed by using artificial bee colony algorithm; population of ABC is initialized, and population size is Each food source represents a path, and contains nodes; integer coding is used for coding of the food source, and the initialization formula of the th food source is: ； wherein, is a honey source the first node in the set is generated according to the following rule: ； wherein, is the set of all nodes in the environment except the start and end nodes; , is the total number of nodes; denotes a random selection of a node from the set of nodes. 3.The robot path planning method based on artificial bee colony and reinforcement learning according to claim 1, wherein, Step S2 specifically comprises the following steps: S2.1, the path length and smoothness are comprehensively considered by using a fitness function, the fitness function The expression is: ； wherein represents the total length of the path, the calculation formula is represented as: ； wherein, denotes the Euclidean distance; denotes the number of inflection points of the path ; and is a smoothness penalty coefficient. S2.2, smoothness is embodied by the number of inflection points, the number of inflection points is determined by calculating the vector angle of three consecutive points, assuming three consecutive points , , , two vectors and are formed, and the cosine value of the angle between the two vectors is: ； If then is determined as a turning point; wherein, is a preset threshold, The greater, the weaker the turning point determination strength; S2.3, based on the fitness function The neighborhood search and population update are performed; the new hive generation formula for the employed bees phase is: ； where, is a new nectar; is a current nectar; is a randomly selected domain nectar; is a random number in the range [-1, 1].

4. The robot path planning method based on artificial bee colony and reinforcement learning according to claim 1, wherein, Step S3 specifically is: When the number of iterations of the ABC algorithm reaches the preset maximum value Or the optimal fitness of the population in a continuous Intra-generation improvement less than threshold When the time is right, terminate the ABC algorithm search; select the population with the highest fitness from the final population. A set of non-repeating paths constitutes the elite solution set. The screening formula is: ； wherein is a fitness threshold, is a first path.

5. A robot path planning system based on artificial bee colony and reinforcement learning, using the method of any one of claims 1-4, characterized in that, Comprise: A population initialization module for initializing a population of artificial bee colony algorithm according to known environment information, a starting point and a target point, wherein each individual in the population represents a feasible path from the starting point to the target point composed of path points; A global search optimization module for iteratively optimizing the population based on a search mechanism of the artificial bee colony algorithm, wherein the quality of an individual is evaluated by a fitness function that comprehensively considers path length and path smoothness, and the population is updated accordingly; An elite solution set construction module for selecting a batch of high-quality solutions from the final population to form an elite solution set when the global search optimization module meets a preset optimization process termination condition; A reinforcement learning initialization module integrating a reinforcement learning algorithm for encoding the elite solution set as prior knowledge of a reinforcement learning agent to initialize a value function, and setting a state space, an action space and a reward function that combines path length and smoothness objectives; A strategy fine-tuning module for running a reinforcement learning algorithm to search and optimize strategies based on the initialized value function to fine-tune the path strategy; A path output module for outputting a decided optimal path after convergence, which meets the requirements of shortest length and fewest turning points.

Images