CN-122018556-A - Formation obstacle avoidance method based on escape algorithm

CN122018556ACN 122018556 ACN122018556 ACN 122018556ACN-122018556-A

Abstract

The invention discloses a formation obstacle avoidance method based on an escape algorithm, which is used for solving the problem of formation cooperative obstacle avoidance in a complex environment. The method comprises the steps of initializing a population and reserving elite individuals, differentially updating the individuals according to panic indexes in an initial iteration stage to ensure search diversity, switching to a development stage after iteration times reach a set threshold value to deep dig an optimal solution, and finally outputting a formation obstacle avoidance path scheme. The method can realize efficient collaborative path planning and adapt to multi-class formation obstacle avoidance scenes.

Inventors

LIU MENGQI
WANG JIN
ZHANG YAN

Assignees

中国船舶集团有限公司第七六〇研究所

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (5)

1. The formation obstacle avoidance method based on the escape algorithm is characterized by comprising the following steps of: s1, initializing a set scale population, describing individuals by multidimensional vectors, limiting values on corresponding upper and lower bounds, evaluating fitness, sequencing in ascending order, and storing the optimal individuals into an elite pool; S2, when the iteration times do not exceed the threshold value, the individuals are classified into three categories, the panic index is calculated, the cool static individuals are updated rationally, the individuals follow and the panic individuals are explored irregularly, and diversified searching is realized; S3, entering a development stage after the iteration times exceed a threshold value, wherein all individuals are regarded as cool and quiet, and optimizing and updating by combining elite pools and random individual positions to deeply mine an optimal solution area; And S4, after the iteration termination condition is met, extracting optimal individual parameters from the elite pool, converting the optimal individual parameters into a formation obstacle avoidance path scheme, and outputting guidance formation to cooperatively avoid the obstacle in a complex environment.
2. The escape algorithm-based formation obstacle avoidance method of claim 1, wherein the specific content of S1 is as follows: Initializing a population of size N, each individual being described by a D-dimensional vector X i ＝(X i,1 ,X i,2 ,…,X i,D ), the ith individual having a value in the jth dimension of X i,j ＝LB j +r i,j ×(UB j -LB j Wherein UB j and LB j are the upper and lower bounds of the j-th dimension, respectively, ensuring that the initial positions of each individual are randomly distributed in the feasible space, r i,j is a random variable uniformly distributed between 0 and 1; after initializing the population, the fitness of each individual is evaluated, the population is then sorted in ascending order according to fitness, and the individuals with the best fitness are stored in elite pool E, and this parameter represents the number of potential safety exits found by the population.
3. The escape algorithm-based formation obstacle avoidance method as set forth in claim 1, wherein the specific content of S2 is: When the number of iterations is not greater than T/2, the individual is classified as cold, compliant or panic, which correspond to different behavioral responses during evacuation, and the panic index is calculated at the beginning of each iteration by the following formula Wherein T is the current iteration number, and T is the total iteration number; The individual in the cool static group shows rationality, and the position update formula is that Wherein X i,j is the value of the ith individual in the j-th dimension, m 1 is a binary variable, ω 1 is an adaptive Levy weight, C j is the mean of the calm group in the j-th dimension, P (t) is the panic index, and vector v c,j is defined as follows v c,j ＝R c,j -X i,j +ε j Where R c,j is the randomly generated position in the cool static group in that dimension, X i,j is the value of the ith individual in the j-th dimension, ε j is a slight adjustment of the individual's movements; individuals who are compliant with the group will follow the behavior of the cool and panic groups and the location update formula is Where X i,j is the value of the ith individual in the j-th dimension, m 1 and m 2 are binary variables, ω 1 and ω 2 are adaptive Levy weights, X p,j is the individual randomly selected from the panic group, P (t) is the panic index, and vector v h,j is defined as follows v h,j ＝R h,j -X i,j +ε j Where R h,j is the randomly generated position in the compliance group in that dimension, X i,j is the value of the ith individual in the j-th dimension, ε j is a slight adjustment of the individual's movements; The panic-driven individual exhibits a more irregular exploration in the solution space with a location update formula of Wherein X i,j is the value of the ith individual in the j-th dimension, m 1 and m 2 are binary variables, ω 1 and ω 2 are adaptive Levy weights, E j is the individual randomly selected from the elite pool, X rand,j is the individual randomly selected from the population, P (t) is the panic index, and vector v h,j is defined as follows v p,j ＝R p,j -X i,j +ε j Where X i,j is the value of the ith individual in the j-th dimension, R p,j is the randomly generated position in the panic group in that dimension, ε j is a slight adjustment of the individual's movements.
4. The escape algorithm-based formation obstacle avoidance method as set forth in claim 1, wherein the specific content of S3 is: as the iteration times exceed T/2, the algorithm enters a development stage, all individuals are regarded as cold and quiet individuals at the moment, and the position information updating formula at the stage is as follows Where X i,j is the position of the ith individual in the j-th dimension, m 1 and m 2 are binary variables, ω 1 and ω 2 are adaptive Levy weights, E j is the position of one member of the elite pool, and X rand,j is the position of an individual randomly selected from the population.
5. The escape algorithm-based formation obstacle avoidance method as set forth in claim 1, wherein the specific content of S4 is: judging whether iteration reaches the preset total times or meets convergence conditions, if so, terminating the iteration, extracting parameter information corresponding to the optimal individual from an elite pool, converting the parameter information into a specific path planning scheme for formation obstacle avoidance, definitely forming the travelling track and obstacle avoidance nodes of each member of the formation, and outputting the scheme to guide the formation to cooperatively avoid the obstacle in a complex environment, so that travelling safety and efficiency are ensured.

Description

Formation obstacle avoidance method based on escape algorithm Technical Field The invention belongs to the technical field of intelligent control and path planning, relates to a multi-agent formation cooperative obstacle avoidance technology, and particularly relates to a formation obstacle avoidance method based on an escape algorithm. Background With the continuous increase of demands of multi-agent systems in the fields of logistics inspection, military reconnaissance, disaster relief and the like, formation cooperation obstacle avoidance becomes a core scene for the multi-agent to execute complex tasks, and higher requirements are provided for the high efficiency, the synergy and the robustness of path planning. The existing formation obstacle avoidance method is mostly based on a traditional group intelligent algorithm or a geometric obstacle avoidance strategy, has the problems that the diversity of a searching stage is insufficient and a developing stage is easy to fall into a local optimal solution, and is difficult to meet the real-time collaborative obstacle avoidance requirement of formation in a complex dynamic environment. The invention provides a formation obstacle avoidance method based on an escape algorithm, which effectively solves the problems of weak cooperativity and low path planning precision of the traditional method through population differentiation updating and deep development guided by an elite pool, and provides a reliable scheme for safe and efficient operation of multi-agent formation in a complex environment. Disclosure of Invention The invention provides a formation obstacle avoidance method based on an escape algorithm for solving the problems in the prior art, and provides the following technical scheme: The method comprises the steps of S1, initializing a set scale population, describing individuals by multidimensional vectors, limiting values on corresponding upper and lower bounds, evaluating ascending order of fitness, storing optimal individuals into an elite pool, S2, dividing the individuals into three types and calculating panic indexes when iteration times do not exceed a threshold value, updating rationality of cool individuals, following the individuals, irregularly exploring the panic individuals to realize diversified searching, S3, entering a development stage after the iteration times exceed the threshold value, regarding all the individuals as cool and quiet, combining the elite pool with random individual positions, optimizing and updating to deeply mine an optimal solution area, S4, extracting optimal individual parameters from the elite pool after iteration termination conditions are met, converting the optimal individual parameters into a formation obstacle avoidance path scheme, and outputting guidance formation to cooperatively avoid obstacles in a complex environment. Preferably, the specific content of S1 is: Initializing a population of size N, each individual being described by a D-dimensional vector X i＝(Xi,1,Xi,2,…,Xi,D), the value of the ith individual in the jth dimension being Xi,j＝LBj+ri,j×(UBj-LBj) Where UB j and LB j are the upper and lower bounds of the j-th dimension, respectively, ensuring that the initial positions of each individual are randomly distributed in the feasible space, and r i,j is a random variable uniformly distributed between 0 and 1. After initializing the population, the fitness of each individual is evaluated, the population is then sorted in ascending order according to fitness, and the individuals with the best fitness are stored in elite pool E, and this parameter represents the number of potential safety exits found by the population. Preferably, the specific content of S2 is: When the number of iterations is not greater than T/2, the individual is classified as cold, compliant or panic, which correspond to different behavioral responses during evacuation, and the panic index is calculated at the beginning of each iteration by the following formula Where T is the current iteration number and T is the total iteration number. The individual in the cool static group shows rationality, and the position update formula is that Wherein X i,j is the value of the ith individual in the j-th dimension, m 1 is a binary variable, ω 1 is an adaptive Levy weight, C j is the mean of the calm group in the j-th dimension, P (t) is the panic index, and vector v c,j is defined as follows vc,j＝Rc,j-Xi,j+εj Where R c,j is the randomly generated position in the cool static group in this dimension, X i,j is the value of the ith individual in the j-th dimension, ε j is a slight adjustment of the individual's movements. Individuals who are compliant with the group will follow the behavior of the cool and panic groups and the location update formula is Where X i,j is the value of the ith individual in the j-th dimension, m 1 and m 2 are binary variables, ω 1 and ω 2 are adaptive Levy weights, X p,j is the individual randomly selected from