A robot path optimization method and related device based on RRT algorithm and particle swarm algorithm

Abstract

The application discloses a robot path optimization method based on an RRT algorithm and a particle swarm algorithm and related devices. In the application, an initial path is generated based on the RRT algorithm, and a key path segment set is determined by using local features of the initial path. Then, the key path segment set is used as a key guide of an initial population, so that the quality and information coordination degree of initial path individuals are improved. Path individuals are optimized in stages based on path fitness, and the search process is divided into three stages of exploration, development and balance, so that the global optimization ability and search diversity in a complex environment are improved, and the method can adapt to different environment complexities and planning requirements, and provides high-quality path planning for robot dynamics.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of robot path planning technology, and in particular to a robot path optimization method based on RRT algorithm and particle swarm algorithm and related devices. BACKGROUND

[0002] With the rapid development of intelligent manufacturing, automation and robot technology, mobile robots are increasingly widely used in logistics, medical treatment and industrial manufacturing and other fields. People's requirements for path planning of mobile robots in complex environments are also increasing. How to achieve efficient, safe and fully considerate path planning of dynamic obstacles in complex environments has become one of the key technologies in the design of modern mobile robot devices. In the existing path planning technology, the Rapidly-exploring Random Tree (RRT) algorithm and its improved RRT algorithm provide a series of highly implementable planning algorithms. This kind of algorithm realizes the search of feasible paths in high-dimensional complex environment by relying on the application of random tree, and has basic mobility. However, the path generated by this kind of algorithm is not optimal in terms of path quality. The RRT algorithm path usually has large repetition, zigzag and very planning, which leads to low robot moving efficiency, large oscillation and increased energy consumption. At the same time, the RRT algorithm has limited response to dynamic obstacles and sudden situations in the environment. Sudden obstacles can easily lead to complete re-planning of the path, resulting in decreased efficiency. SUMMARY

[0003] The present application provides a robot path optimization method based on RRT algorithm and particle swarm algorithm and related devices, which is used to solve the technical problem of poor path quality generated by the existing robot path planning method in complex environment.

[0004] The present application provides a robot path optimization method based on RRT algorithm and particle swarm algorithm, which comprises:

[0005] Obtaining a target path task of a robot, generating an initial path based on the RRT algorithm according to the target path task;

[0006] Extracting the local feature vectors of each path segment in the initial path, and comprehensively scoring and analyzing the local feature vectors of each path segment to determine a set of key path segments;

[0007] Generating a plurality of path individuals according to the set of key path segments based on Gaussian disturbance, and calculating the path fitness of each path individual; sorting the path individuals in descending order of path fitness to obtain a particle swarm in a plurality of fitness intervals;

[0008] The particle swarm in different fitness intervals is optimized in a corresponding exploration stage, development stage and balance stage, new path individuals are obtained by optimization, and path fitness of the new path individuals is calculated; the new path individuals are sorted in descending order according to path fitness, and new particle swarms in multiple fitness intervals are obtained;

[0009] When the number of iterations does not reach the predetermined number of cycles, the step of optimizing the particle swarm in different fitness intervals in a corresponding exploration stage, development stage and balance stage is executed; when the number of iterations reaches the preset threshold, the path individual with the optimal path fitness in the entire iteration process is selected as the target path individual and output.

[0010] Optionally, the step of extracting the local feature vectors of each path segment in the initial path, and comprehensively analyzing the local feature vectors of each path segment to determine the key path segment set, comprises:

[0011] Extracting the segment length feature, curvature feature and obstacle distance feature of each path segment in the initial path;

[0012] Comprehensively analyzing the segment length feature, curvature feature and obstacle distance feature of each path segment based on a preset comprehensive scoring function to obtain a scoring result of each path segment;

[0013] Sorting all the path segments in ascending order according to the corresponding scoring results, selecting the first Z path segments as key path segments, and generating a key path segment set based on all the key path segments; wherein Z is a positive integer.

[0014] Optionally, the step of generating a plurality of path individuals based on the key path segment set Gaussian disturbance, and calculating the path fitness of each path individual; sorting the path individuals in descending order according to the path fitness to obtain a particle swarm in multiple fitness intervals, comprises:

[0015] Taking the midpoint of each key path segment in the key path segment set as an individual generation center;

[0016] Sampling the path individuals of each key path segment around the individual generation center corresponding to each key path segment by taking the segment length of each key path segment as a Gaussian disturbance standard deviation;

[0017] Calculating the fitness of each path individual based on a preset evaluation function to obtain the path fitness of each path individual;

[0018] The path individuals are sorted in descending order of path fitness, the sorted path individuals are divided based on a preset division ratio, and particle groups corresponding to a first fitness interval, a second fitness interval and a third fitness interval are obtained.

[0019] Optionally, the optimization of the particle groups in different fitness intervals in the corresponding exploration stage, development stage and balance stage, and the calculation of the path fitness of the new path individuals, comprises:

[0020] The dynamic adjustment operator, the adjustment speed and the weight factor for search exploration are calculated according to the current iteration number and the predetermined cycle number;

[0021] The particle groups in the first fitness interval are updated and optimized in the exploration stage based on the dynamic adjustment operator, and new path individuals are obtained through optimization;

[0022] The particle groups in the second fitness interval are updated and optimized in the development stage based on the adjustment speed, and new path individuals are obtained through optimization;

[0023] The particle groups in the third fitness interval are updated and optimized in the balance stage based on the weight factor, and new path individuals are obtained through optimization;

[0024] The path fitness of each new path individual is calculated.

[0025] Optionally, in the exploration stage, when the dynamic adjustment operator is greater than a preset adjustment operator, the update and optimization mode of the new path individual is:

[0026]

[0027]

[0028] In the formula: represents the position vector of the differential random solution individual, represents the mutation factor, and represents the position vector of the two path individuals before updating, which are randomly sampled from the particle groups in the first fitness interval; represents the position vector of the Kth path individual after updating, represents the position adjustment factor, represents the position vector of the kth path individual before updating in the particle groups in the first fitness interval, represents a random number between 0 and 1, represents a random number vector conforming to the standard normal distribution; round(*) is the rounding function.

[0029] When the dynamic adjustment operator is less than or equal to the preset adjustment operator, the optimization updating mode of the new path individual is:

[0030]

[0031] In the formula: represents a local updating position operation, represents a dimension index; represents a random integer in the interval [1, D], and D represents the path dimension.

[0032] Optionally, the optimization updating process in the development stage includes:

[0033]

[0034] In the formula: represents a dynamic iteration factor, represents an adjustment speed, represents a dynamic scaling factor.

[0035] Optionally, the differential optimization updating process in the balance stage includes:

[0036] When , the updating mode of the new path individual is represented as:

[0037]

[0038] In the formula: represents a weight factor, represents the position vector of the path individual in the optimal position in the particle swarm of the current three fitness intervals, represents a uniform random number matrix of 1 row and D columns, represents a spiral trajectory adjustment factor, represents a dynamic disturbance factor for adjusting the updating of the particle position, represents a dynamic step adjustment factor, which is used to adaptively adjust the search step according to the iteration number in the optimization process, and when , the updating mode of the new path individual is represented as:

[0039]

[0040] In the formula: represents a random initialization individual operation; represents a uniform random number matrix of N rows and D columns, and the element value range is [0, 1];

[0041] When , the updating mode of the new path individual is represented as:

[0042]

[0043] wherein: represents a displacement operator.

[0044] The application also provides a computer device comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the steps of the robot path optimization method according to any one of the above.

[0045] The application also provides a computer readable storage medium having stored thereon a computer program / instruction, which, when executed by a processor, implements the steps of the robot path optimization method according to any one of the above.

[0046] The application also provides a computer program product comprising a computer program / instruction, which, when executed by a processor, implements the steps of the robot path optimization method according to any one of the above.

[0047] As can be seen from the above technical solutions, the application has the following advantages:

[0048] The application provides a robot path optimization method and related device based on an RRT algorithm and a particle swarm algorithm, and the method comprises the following steps: obtaining a target path task of a robot, generating an initial path according to the target path task based on the RRT algorithm; extracting local feature vectors of each path segment in the initial path, and comprehensively scoring and analyzing the local feature vectors of each path segment to determine a key path segment set; generating a plurality of path individuals according to the key path segment set Gaussian disturbance, and calculating path fitness of each path individual; sorting the path individuals in descending order of path fitness to obtain a particle swarm in a plurality of fitness intervals; optimizing the particle swarm in different fitness intervals in a corresponding exploration stage, development stage and balance stage, obtaining new path individuals and calculating path fitness of the new path individuals; sorting the new path individuals in descending order of path fitness to obtain a new particle swarm in a plurality of fitness intervals; when the number of iterations does not reach a predetermined number of cycles, jumping to the step of optimizing the particle swarm in different fitness intervals in a corresponding exploration stage, development stage and balance stage, obtaining new path individuals and calculating path fitness of the new path individuals; when the number of iterations reaches a preset threshold, selecting a path individual with optimal path fitness in the entire iteration process as a target path individual and outputting the target path individual.

[0049] In the application, the key path segment set is used as the key guide of the initial population, the quality and information coordination degree of the initial path individual are improved, and the path individual is optimized in stages based on the path fitness, the search process is divided into three stages of exploration, development and balance, the global optimization ability and search diversity in the complex environment are improved, the different environment complexity and planning requirements are adapted, the high-quality path planning is provided for the robot dynamics, and thus the technical problem of poor path quality generated by the existing robot path planning method in the complex environment is solved. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0051] Figure 1 A step flow chart of a robot path optimization method based on the RRT algorithm and the particle swarm algorithm is provided for the embodiments of the present application.

[0052] Figure 2 An execution flow chart of a robot path optimization method based on the RRT algorithm and the particle swarm algorithm is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0053] The embodiments of the present application provide a robot path optimization method and related device based on the RRT algorithm and the particle swarm algorithm, which are used to solve the technical problem of poor path quality generated by the existing robot path planning method in the complex environment.

[0054] In order to make the application purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the following described embodiments are only some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0055] The method provided in the present application relates to the concept of path optimization. In order to facilitate understanding, the related concepts of path optimization involved will be introduced.

[0056] Rapidly-exploring Random Tree (RRT): A probabilistic algorithm commonly used in robot motion planning and path searching, belonging to the category of directed graph search methods. Its core idea is to construct a constantly expanding tree structure in the state space through random sampling, gradually exploring unknown space, and finally finding a feasible path from the starting point to the target point; its improved version RRT* algorithm introduces rewiring mechanism to optimize path cost and ensure asymptotic optimality.

[0057] Gaussian Perturbation-based Self-Organizing Multi-stage Optimization (GPSOM): An intelligent optimization algorithm that combines the ideas of Particle Swarm Optimization (PSO) and improves through Gaussian perturbation and self-organizing multi-stage mechanism; Gaussian perturbation is usually combined with techniques such as "simulated annealing" and "mutation operator" in optimization algorithms, while the self-organizing multi-stage mechanism draws on the stage adaptation strategy in biological evolution. This algorithm combines mathematical models (Gaussian distribution) and logical controls (stage switching conditions) to outperform traditional PSO in solving high-dimensional, multi-peak optimization problems.

[0058] Please refer to Figure 1 and Figure 2 , the present application provides a robot path optimization method based on RRT algorithm and particle swarm algorithm, the method comprising:

[0059] Step 101, obtain the target path task of the robot, and generate an initial path based on the RRT algorithm according to the target path task.

[0060] In this embodiment, the target path task of the robot includes the path starting point and terminal and obstacle information in the path. This step uses an improved RRT algorithm (such as Informed RRT*-Connect algorithm) to generate an initial path from the path starting point to the path terminal that can reach without colliding with obstacles, the initial path is composed of multiple discrete path points (waypoints), ensuring the connectivity and feasibility of the path, meeting the planning space constraints and safety requirements.

[0061] Each discrete path point includes two-dimensional or three-dimensional coordinate information, the overall dimension D of the initial path is determined by the number of path discrete points and the coordinate dimension, for example, for n path discrete points in a two-dimensional space, D = 2n. Then an initial path P from the path starting point to the terminal can be represented as, wherein Represents the i-th path discrete point of the initial path P.

[0062] Step 102 : extracting the local feature vector of each path segment in the initial path, and performing comprehensive scoring analysis on the local feature vector of each path segment to determine a critical path segment set.

[0063] This step specifically includes:

[0064] Sub-step 1021 : extracting the segment length feature, curvature feature, and obstacle distance feature of each path segment in the initial path.

[0065] In this embodiment, the segment length feature of each path segment in the initial path , curvature characteristics and obstacle distance characteristics The specific extraction method is:

[0066] Calculate segment length feature by Euclidean length of path segment ,Right now:

[0067]

[0068] Where: Indicates the length of the i-th path segment.

[0069] Calculate the local curvature of a path segment using a three-point approximation ,Right now:

[0070]

[0071] Where: represents the local curvature of the i-th path segment, Represents discrete points on the path The local turning angle.

[0072] Path discrete points Obstacle distance feature to the nearest obstacle Expressed as:

[0073]

[0074] Where, represents the obstacle distance of the i-th path segment, Represents a collection of obstacle information.

[0075] The above features together constitute the local feature vector of the corresponding path segment :

[0076]

[0077] Where: local feature vector of the i-th path segment.

[0078] In the embodiment, for each path segment of the initial path, three key structural features of the path segment are calculated, including a segment length, a local curvature and a distance to an obstacle; the segment length is calculated by using the Euclidean distance, the curvature is approximated by using a three-point method, and the distance to the obstacle is obtained by measuring a minimum distance from a node to a nearest obstacle boundary, which comprehensively reflect smoothness, safety and rationality of each path segment.

[0079] In substep 1022, the segment length feature, the curvature feature and the obstacle distance feature of each path segment are comprehensively scored and analyzed based on a preset comprehensive scoring function, to obtain a scoring result of each path segment.

[0080] It can be understood that the preset comprehensive scoring function is used to evaluate the structure of the path segment, and the higher the scoring result of the path segment is, the better the structure of the path segment is, that is, the path segment is more smooth, farther away from the obstacle and more reasonable in path length.

[0081] The preset comprehensive scoring function is represented as:

[0082]

[0083] In the formula, scorei represents the scoring result of the i-th path segment, are weighted feature coefficients, is a small constant for preventing division by zero, and Lavg represents an average path length of the initial path.

[0084] In substep 1023, all path segments are sorted in ascending order according to the corresponding scoring results, the first Z path segments are selected as key path segments, and a key path segment set is generated based on all the key path segments; Z is a positive integer.

[0085] In step 103, a plurality of path individuals are generated according to the key path segment set and Gaussian disturbance, and path fitness of each path individual is calculated; the path individuals are sorted in descending order according to the path fitness, to obtain a particle swarm of a plurality of fitness intervals.

[0086] This step specifically includes:

[0087] In substep 1031, the midpoint of each key path segment in the key path segment set is taken as an individual generation center.

[0088] In the traditional GPSOM algorithm, an initial population is randomly generated by Gaussian disturbance on a basic path, while in the method, the key path segment information is used to guide the population generation process, so that the population individuals are concentrated in a potential optimal solution region. ​​​

[0089] wherein the midpoint of the critical path segment of the ith critical path segment is denoted as:

[0090]

[0091] In substep 1032, the path individual of each critical path segment is generated around the center of the corresponding individual by taking the segment length of each critical path segment as the standard deviation of Gaussian disturbance.

[0092] It can be understood that based on the midpoint position of the critical path segment, the Gaussian disturbance mechanism is adopted, and the disturbance intensity is taken as the standard deviation of the path segment length, which ensures that the initial solution is concentrated in the excellent structure area, thereby improving the convergence speed and solution quality of the swarm intelligence optimization algorithm.

[0093] The path individual corresponding to each critical path segment is denoted as:

[0094]

[0095] In the formula: denotes the position vector of the path individual generated by the ith critical path segment, denotes the disturbance intensity factor, denotes is the unit matrix.

[0096] By sequentially connecting the disturbance points of each segment by performing the above sampling process on multiple critical path segments, a complete path individual can be constructed. By repeatedly performing the process, multiple initial path individuals with high quality and concentrated distribution in the potential optimal solution area can be generated to form the initial population of the optimization algorithm , .

[0097] In view of the randomness of the initial population generated by the traditional optimization algorithm and the insufficient global search ability, the present application introduces a guided population initialization strategy based on the structural characteristics of the path nodes. Under this strategy, the length, local curvature and minimum distance of the nodes to the obstacles of each segment in the path are finely extracted to construct a comprehensive scoring function combining smoothness, safety distance and reasonable length of the path. The function adjusts the contribution of each feature with a weight parameter, scientifically selects the critical path segment with the highest score, and then takes the midpoint of the critical segment as the center point of the generated individual, uses the segment length as the standard deviation of Gaussian disturbance, combines the disturbance intensity factor and the unit matrix, and realizes the Gaussian disturbance sampling of the initial population. This strategy effectively improves the quality and information coordination degree of the initial solution of the population, and can significantly enhance the focusing ability on the optimal solution area.

[0098] In substep 1033, the fitness of each path individual is calculated based on the preset evaluation function to obtain the path fitness of each path individual.

[0099] wherein the preset evaluation function is represented as:

[0100]

[0101] wherein: represents the path fitness of the i-th path individual, represents the total length of the path of the i-th path individual, and both represent penalty coefficients, represents the number of collisions between the path of the i-th path individual and the obstacles, represents the overall curvature (e.g., square of the turning angle) of the path of the i-th path individual.

[0102] The preset evaluation function constructed in the embodiment comprehensively considers the path length, collision risk and curvature smoothness, and can achieve dynamic trade-off among safety, smoothness and efficiency of the path.

[0103] In step 1034, the path individuals are sorted in descending order of path fitness, and the sorted path individuals are divided based on a preset division ratio to obtain particle groups corresponding to the first fitness interval, the second fitness interval and the third fitness interval.

[0104] In the initial stage, the path individuals are sorted in descending order of path fitness, and the path individuals of the initial population are divided into three stages for exploration. Specifically, the path individuals in the middle 30%-40% of the path fitness sorting are taken as the particle group of the first fitness interval, the path individuals in the front 20%-30% of the path fitness sorting are taken as the particle group of the second fitness interval, and the path individuals in the middle 30%-40% of the path fitness sorting are taken as the particle group corresponding to the third fitness interval.

[0105] In step 104, the particle groups in different fitness intervals are optimized in the corresponding exploration stage, development stage and balance stage, new path individuals are obtained by optimization, and the path fitness of the new path individuals is calculated. The new path individuals are sorted in descending order of path fitness to obtain particle groups of new multiple fitness intervals.

[0106] It should be noted that the present application combines a Gaussian perturbation particle swarm optimization algorithm (GPSOM), divides the path individuals of the initial population into three stages for exploration, and divides the search process into three stages of exploration, development and balance. The stage division mechanism can effectively alleviate the defect that the traditional particle swarm algorithm is easy to fall into local optimum, and improve the global optimization ability and search diversity.

[0107] In this step, the particle swarm in different fitness intervals is optimized in the corresponding exploration stage, development stage and balance stage, and the process of obtaining new path individuals and calculating the path fitness of the new path individuals includes:

[0108] Step S11, according to the current iteration number and the predetermined cycle number, the dynamic adjustment operator, the adjustment speed and the weight factor of the search exploration are calculated.

[0109] It can be understood that the dynamic adjustment operator, the Sine-Cosine speed adjustment and the weight factor are used to adjust the search progress of the exploration stage, the development stage and the balance stage in turn.

[0110] The calculation method of the dynamic adjustment operator is as follows:

[0111]

[0112] In the formula: represents a random number in [0, 1], represents a step factor that decreases with the iteration number, represents the current iteration number, represents the predetermined cycle number.

[0113] The calculation method of the Sine-Cosine speed adjustment is as follows:

[0114]

[0115]

[0116]

[0117] In the formula: represents a nonlinear speed adjustment factor; represents a uniform random number matrix of 1 row and D columns, and the element value range is [0, 1]; and respectively represent the upper and lower boundaries of adjustment; denotes a perturbation term, which is dynamically adjusted as the iteration proceeds.

[0118] where the weight factor is calculated as:

[0119]

[0120] Step S12, the particle swarm in the first fitness interval is updated by the dynamic adjustment operator to perform the differentiated optimization in the exploration stage, and a new path individual is obtained.

[0121] In the embodiment, the particles with poor fitness are jumped out of the local optimal search according to the dynamic adjustment operator in the exploration stage, and the global search ability is enhanced by using the step-down step factor and the random perturbation mechanism. In order to save the calculation complexity, the preset adjustment operator can be set to 1.

[0122] When the dynamic adjustment operator is greater than 1, the updating method of the new path individual includes:

[0123]

[0124]

[0125]

[0126]

[0127] In the formula, position vector of the difference random solution individual, position of the two path individuals before updating, mutation factor, and position of the two path individuals before updating, position of the Kth path individual after updating, position adjustment factor, position vector of the kth path individual before updating, random number vector conforming to the standard normal distribution (mean value is 0 and variance is 1), decay factor, which decreases with the iteration, constraint factor, which is initialized as a zero vector, natural logarithm base; round(*) is the rounding function (rounded to the nearest integer).

[0128] When the dynamic adjustment operator is less than or equal to 1, the updating method of the new path individual includes:

[0129]

[0130]

[0131]

[0132] wherein: represents a local update position operation, is a dimension index; represents a random integer within the interval [1, D]; represents an iteration control factor, which decreases with iteration and is multiplied by a normal random disturbance; represents the first row of a diagonal matrix extraction.

[0133] Step S13, based on the adjustment speed, the particle swarm in the second fitness interval is updated for optimization in the development stage, and a new path individual is obtained through optimization.

[0134] In this embodiment, the development stage focuses on fine-tuning elite particles with high fitness, and by means of nonlinear Sine-Cosine speed adjustment and dynamic scaling factor, the local search precision and path smoothness are improved.

[0135] In the development stage, the update method of the new path individual includes:

[0136]

[0137]

[0138]

[0139]

[0140] wherein: represents a dynamic iteration factor, represents a dynamic scaling factor; represents an adjustment factor that changes with iteration.

[0141] Step S14, based on the weight factor, the particle swarm in the third fitness interval is updated for differential optimization in the balance stage, and a new path individual is obtained through optimization.

[0142] In this embodiment, the balance stage dynamically switches global and local search strategies through the weight factor, maintains population diversity, and prevents premature convergence.

[0143] When , the update method of the new path individual is represented as:

[0144]

[0145]

[0146]

[0147] In the formula: represents the position vector of the path individual in the optimal position in the particle group of the current three fitness intervals, represents a dynamic disturbance factor for adjusting the particle position update, represents a dynamic step adjustment factor for adaptively adjusting the search step according to the iteration number in the optimization process. represents a spiral trajectory adjustment factor; M represents the maximum number of iterations, and the value range is [t, T].

[0148] When , the update mode of the new path individual is represented as:

[0149]

[0150]

[0151] In the formula: represents a random initialization individual operation for enhancing the particle diversity of the particle group; represents a uniform random number matrix of N rows and D columns, and the element value range is [0, 1], wherein N represents the scene dimension, and when the path is located in a two-dimensional space, N takes a value of 2, and when the path is located in a three-dimensional space, N takes a value of 3.

[0152] When , the update mode of the new path individual is represented as:

[0153]

[0154]

[0155] In the formula: represents a displacement operator for controlling the degree of movement to the optimal solution.

[0156] Step S15, the path fitness of each new path individual is calculated.

[0157] After each round of iteration, the path fitness of the new path individual needs to be calculated and the population is updated, that is, the new path individual is sorted in descending order according to the path fitness, and the new particle group of multiple fitness intervals is updated.

[0158] ​Step 105, when the number of iterations does not reach the predetermined number of cycles, jump to the step of performing the optimization of the corresponding exploration stage, development stage and balance stage of the particle group in different fitness intervals, and the new path individual is obtained by optimization, and the path fitness of the new path individual is calculated; when the number of iterations reaches the preset threshold, the path individual with the optimal path fitness in the entire iteration process is selected as the target path individual and output.

[0159] The fusion RRT algorithm and GPSOM algorithm, the Gaussian disturbance population initialization strategy based on path node characteristics is designed, the optimization population is effectively guided to the high potential area distribution, the synergistic effect between path information and path optimization is enhanced, and the quality and optimization efficiency of the path of the multi-degree-of-freedom robot in the dynamic complex environment are significantly improved. In the iterative optimization process, a plurality of dynamic adjustment parameters such as disturbance intensity factor, step length reduction factor and weight adjustment coefficient are introduced, the flexible regulation of the algorithm in different iteration stages is realized, different environment complexity and planning requirements are adapted, the fitness of all new solutions is calculated in real time and the stage division is dynamically adjusted, the global optimal particle is reserved, and stable convergence of the algorithm is ensured, and high-quality path planning is provided for dynamic robot.

[0160] The application is also compatible with two-dimensional and three-dimensional space planning, and is suitable for path planning tasks of multi-degree-of-freedom robots in indoor and outdoor complex environments with multiple obstacles. Overall, the application realizes efficient fusion of information in the path generation and optimization process, significantly improves the accuracy, robustness and calculation efficiency of path planning, and provides an optimization technology for autonomous navigation of robots in dynamic environments.

[0161] The robot path optimization method based on the RRT algorithm and the particle swarm algorithm has the following advantages:

[0162] 1. The fast exploration of complex configuration space of the RRT algorithm and the global optimization ability of the GPSOM algorithm are combined, the effective coupling of the path planning result and the swarm intelligence algorithm is realized, the distribution quality of the initial solution is improved, and fast and fine path planning is provided.

[0163] 2. The population initialization guided by the path structure characteristics, the comprehensive scoring function based on the path segment structure characteristics, the scientific quantification of path advantages and disadvantages, the accurate guidance basis for optimization, the dynamic adjustment of the Gaussian disturbance intensity based on the path structure characteristics, the population individuals closer to the potential optimal solution area, the optimization efficiency and the diversity of solutions, the convergence speed and stability of the optimization algorithm are improved, and the invalid search is reduced.

[0164] 3. The initial population is iteratively optimized based on the GPSOM algorithm, phased exploration is realized, global exploration and local development are coordinated, the smoothness of the path is gradually improved, the problems of falling into local optimum and path unsmoothness of the traditional path planning algorithm when used alone are effectively solved, the generated path is smoother and farther away from obstacles, and the path is adapted to complex dynamic environments.

[0165] The application further provides a computer device, including a memory, a processor and a computer program stored on the memory, the processor executes the computer program to realize the steps of the robot path optimization method according to any one of the above.

[0166] The application further provides a computer readable storage medium, which stores a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the robot path optimization method according to any one of the above.

[0167] The application further provides a computer program product, which includes a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the robot path optimization method according to any one of the above.

[0168] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0169] In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division manner for actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0170] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, can be located in one place, or can be distributed on a plurality of network units. According to actual needs, some or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0171] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0172] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0173] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or modules does not have to be limited to only those steps or modules clearly listed, but can include other steps or modules that are not clearly listed or inherent to the process, method, product or device. The naming or numbering of the steps appearing in the present application does not mean that the steps in the method flow must be executed in the order / time sequence indicated by the naming or numbering. The flow steps that have been named or numbered can change the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0174] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A robot path optimization method based on RRT algorithm and particle swarm algorithm, characterized in that: The method comprises: Obtaining a target path task of the robot, and generating an initial path according to the target path task based on the RRT algorithm; Extracting local feature vectors of each path segment in the initial path, and performing comprehensive scoring analysis on the local feature vectors of each path segment to determine a critical path segment set; Generate multiple path individuals according to the Gaussian perturbation of the critical path segment set, and calculate the path fitness of each path individual; sort the path individuals in descending order according to the path fitness to obtain particle swarms in multiple fitness intervals; The particle swarms in different fitness intervals are optimized in the corresponding exploration phase, development phase and equilibrium phase to obtain new path individuals and calculate the path fitness of the new path individuals; the new path individuals are sorted in descending order according to the path fitness to obtain new particle swarms in multiple fitness intervals; When the number of iterations does not reach the predetermined number of cycles, the process jumps to the step of optimizing the particle swarms in different fitness intervals in the corresponding exploration phase, development phase, and equilibrium phase, optimizing to obtain new path individuals and calculating the path fitness of the new path individuals; when the number of iterations reaches the preset threshold, the path individual with the best path fitness in the entire iterative process is selected as the target path individual and output.

2. The robot path optimization method according to claim 1, characterized in that: The step of extracting the local feature vector of each path segment in the initial path, performing comprehensive scoring analysis on the local feature vector of each path segment, and determining a critical path segment set includes: Extracting segment length features, curvature features, and obstacle distance features of each path segment in the initial path; Performing a comprehensive scoring analysis on the segment length characteristics, curvature characteristics, and obstacle distance characteristics of each path segment based on a preset comprehensive scoring function to obtain a scoring result for each path segment; All the path segments are sorted in ascending order according to the corresponding scoring results, the first Z path segments are selected as critical path segments, and a critical path segment set is generated based on all the critical path segments; where Z is a positive integer.

3. The robot path optimization method according to claim 2, characterized in that: generating a plurality of path individuals according to the Gaussian perturbation of the critical path segment set, and calculating the path fitness of each of the path individuals; The step of sorting the path individuals in descending order according to the path fitness to obtain particle swarms in multiple fitness intervals includes: Taking the midpoint of each key path segment in the key path segment set as the individual generation center; By taking the segment length of each critical path segment as the Gaussian perturbation standard deviation, sampling around the individual generation center corresponding to each critical path segment obtains the path individuals of each critical path segment; Calculating the fitness of each path individual based on a preset evaluation function to obtain the path fitness of each path individual; The path individuals are sorted in descending order according to the path fitness, and the sorted path individuals are divided based on a preset division ratio to obtain particle swarms corresponding to the first fitness interval, the second fitness interval, and the third fitness interval.

4. The robot path optimization method according to claim 3, characterized in that: The step of optimizing the particle swarms in different fitness intervals in the corresponding exploration phase, development phase, and equilibrium phase, optimizing to obtain new path individuals, and calculating the path fitness of the new path individuals includes: Calculate the dynamic adjustment operator, adjustment speed and weight factor of the search according to the current number of iterations and the predetermined number of cycles; Performing differential optimization updates on the particle swarm in the first fitness interval in the exploration phase based on the dynamic adjustment operator to obtain new path individuals; performing an optimization update on the particle swarm in the second fitness interval in a development phase based on the adjustment speed to obtain a new path individual; Based on the weight factor, the particle swarm in the third fitness interval is subjected to differential optimization and update in the equilibrium phase to obtain a new path individual; Calculate the path fitness of each of the new path individuals.

5. The robot path optimization method according to claim 4, characterized in that: In the exploration phase, when the dynamic adjustment operator is greater than the preset adjustment operator, the optimization update method of the new path individual is: Where: represents the position vector of the individual differential random solution, represents the factor of variation, and Represents the position vectors of two path individuals before the update by random sampling from the particle swarm in the first fitness interval; represents the position vector of the Kth path individual after update, represents the position adjustment factor, Represents the position vector of the k-th path individual before the update in the particle swarm of the first fitness interval, represents a random number between [0,1], Represents a random number vector that conforms to the standard normal distribution, and round(*) is a rounding function; When the dynamic adjustment operator is less than or equal to the preset adjustment operator, the optimization update method of the new path individual is: Where: Represents a local update position operation, Represents dimension index; Represents a random integer in the interval [1, D], where D represents the path dimension.

6. The robot path optimization method according to claim 4, characterized in that: The optimization and update process during the development phase includes: Where: represents the dynamic iteration factor, Indicates the adjustment speed. Indicates the dynamic scaling factor.

7. The robot path optimization method according to claim 4, characterized in that: The differentiated optimization update process in the balancing phase includes: when When , the update method of the new path individual is expressed as: Where: represents the weight factor, Represents the position vector of the path individual in the optimal position in the particle swarm of the current three fitness intervals, represents a uniform random number matrix with 1 row and D columns, represents the spiral trajectory adjustment factor, represents the dynamic perturbation factor used to adjust the particle position update; Represents the dynamic step size adjustment factor, which is used to adaptively adjust the search step size according to the number of iterations during the optimization process; when When , the update method of the new path individual is expressed as: Where: Represents random initialization of individual operations; Represents a uniform random number matrix with N rows and D columns, whose element values ​​range from [0,1]; when When , the update method of the new path individual is expressed as: Where: represents the displacement operator.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the robot path optimization method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the robot path optimization method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the robot path optimization method according to any one of claims 1 to 7 are implemented.

Images