A path planning method based on elephant trunk optimization strategy

Abstract

The application discloses a path planning method based on elephant trunk fish optimization strategy, which comprises the following steps: firstly, generating an initial environment map, initializing the iteration times and N elephant trunk fishes, and calculating the fitness of each elephant trunk fish; secondly, saving the elephant trunk fish with the minimum fitness as the optimal elephant trunk fish, detecting and generating sampling points for each elephant trunk fish, calculating the fitness of the sampling points, and if the fitness is better than the current position, the elephant trunk fish will move to the sampling point, if no sampling point with better fitness is found, the elephant trunk fish will move to the optimal position of the population or a random position in the solution space according to a random probability; then, calculating the fitness of the elephant trunk fish after updating, iterating, taking the optimal elephant trunk fish as the global optimal path, and finally converting the generated path into navigation instructions. The application can quickly find the optimal or nearly optimal path in a complex environment, effectively explore the search space, and avoid the local optimal solution trap.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of path planning of unmanned systems, in particular to a method for path planning using elephant trunk optimization strategy. BACKGROUND

[0002] In many application fields such as unmanned aerial vehicles, unmanned surface vehicles, autonomous driving vehicles and logistics management, path planning is an important problem. Path planning involves finding an optimal path from a starting point to a target point while considering various constraints and obstacles in the environment. The optimized path can minimize travel distance, avoid obstacles, and satisfy other additional constraints such as avoiding restricted areas. In order to achieve efficient and safe navigation, it is necessary to find a high-quality path.

[0003] Traditional path planning methods are often difficult to find optimal or near-optimal paths in large-scale and complex environments due to the increasing complexity of the problem and the dimensionality of the search space. For example, when using differential evolution algorithm for path planning, the search is mainly performed through mutation and crossover operations of the population, which is easily attracted to local optimal solutions and trapped in local optimal solutions. When using ant colony algorithm for path planning, the path selection is gradually adjusted through pheromone tracking and updating, which requires a certain number of iterations and usually takes a long time to converge to the optimal solution. Therefore, a more efficient and flexible algorithm is needed to solve the path planning problem. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide a path planning method based on elephant trunk optimization strategy, which simulates the foraging behavior and group cooperation mechanism of elephant trunk. By using elephant trunk algorithm, the optimal or near-optimal path can be quickly and accurately found in complex environments. The technical solution of the present application is as follows:

[0005] A path planning method based on elephant trunk optimization strategy, comprising the following steps:

[0006] S1, start the sensor system to scan the environment and generate an initial environment map, and after preprocessing, generate a final environment map including the starting point, the ending point and the obstacle circle.

[0007] S2, initialize the iteration number t = 1, initialize N elephant trunks according to the Levy flight formula, and each elephant trunk represents a path; calculate the fitness of each elephant trunk, wherein the fitness is related to the path length and the cost of passing through obstacles.

[0008] S3, save the elephant trunk with the minimum fitness among the N elephant trunks as the optimal elephant trunk.

[0009] S4, for each elephant trunk, perform radar detection to generate sampling points and calculate the fitness of the sampling points.

[0010] S5, for each elephant trunk fish, if a sample point with better fitness than the current position is found, a greedy selection strategy is adopted, and the elephant trunk fish moves to the sample point.

[0011] S6, for each elephant trunk fish, if a sample point with better fitness than the current position is found, a greedy selection strategy is adopted, and the elephant trunk fish moves to the sample point.

[0012] S7, the fitness of each elephant trunk fish after updating is calculated.

[0013] S8, t=t+1, judge whether t reaches the maximum iteration number, if not, return to S2 to perform the next iteration operation, if yes, the final optimal elephant trunk fish is taken as the global optimal path.

[0014] S9, the generated path is converted into navigation instructions to guide the robot to navigate in the environment.

[0015] The present application has the following characteristics and beneficial effects:

[0016] The elephant trunk fish algorithm for path planning has the following effects and advantages: strong robustness and adaptability, can quickly find the optimal or near-optimal path in complex environment; by simulating the search behavior and group cooperation mechanism of elephant trunk fish, the search space can be effectively explored, and the local optimal solution trap can be avoided in the search process; the algorithm parameters and stopping criteria can be flexibly adjusted to meet the needs of different application scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0017] 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 prior art description will be briefly introduced as follows, 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.

[0018] Figure 1 The flowchart of the present application;

[0019] Figure 2 The obstacle circumscribed circle inflation schematic diagram;

[0020] Figure 3 The two-dimensional map path planning result instance diagram of the present application. DETAILED DESCRIPTION

[0021] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0022] In the description of the present application, it needs to be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element indicated must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" and the like are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" and the like can be explicitly or implicitly included one or more. In the description of the present application, unless otherwise stated, the meaning of "a plurality of" is two or more.

[0023] In the description of the present application, it needs to be understood that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood through specific circumstances.

[0024] The present application provides a method for path planning by using elephant trunk fish optimization strategy, as shown in Figure 1 The specific method is:

[0025] S1, data acquisition and pretreatment: starting the sensor system, acquiring the data generated by the sensors such as laser radar and camera, and generating an initial environment map.

[0026] S2, pretreatment: pretreating the initial environment map, as shown in Figure 2 The circumscribed circle of all obstacles in the initial environment map is expanded.

[0027] S3, generating a map: using the pretreated data to construct a feature map representing the environment, the feature map including a starting point, an ending point and an obstacle circle. The subsequent steps are all carried out in this map space.

[0028] S4, designing fitness function: the fitness function considers two components of path length and obstacle cost, and calculates the fitness according to the following formula:

[0029] Path length:

[0030]

[0031] where D represents the number of path key points, d(x i ,x i+1 ) represents the path segment length between key point x i and key point x i+1 .

[0032] Obstacle cost:

[0033]

[0034] where a is the cost coefficient of the obstacle, p(r) is the obstacle cost of the path passing through key point A, R is the circumscribed circle radius of the obstacle, and r is the distance of key point A from the center of the circle.

[0035] Fitness function:

[0036]

[0037] where o is the number of obstacles, m and n are the number of key points and path sampling points respectively, and r i,j represents the distance of the jth key point or sampling point of the path from the center of the circumscribed circle of the ith obstacle.

[0038] S5, path planning: as shown in the figure, the method for path planning using elephant trunk fish optimization strategy of the embodiment of the application includes the following steps: Figure 1 SS1, initialize the iteration number t = 1, initialize N elephant trunk fish according to the Levy flight formula, and one elephant trunk fish represents a path. The initial elephant trunk fish individual is generated according to the following formula:

[0039]

[0040] where X i+1 is the newly generated elephant trunk fish, X i is the elephant trunk fish that generates X i+1 , the first elephant trunk fish of the population is randomly generated in the solution space, alpha is the step scaling factor, and s is the step length obeying the Levy distribution, which is calculated according to the following formula:

[0041]

[0042]

[0043] where beta is usually 1.5, and

[0044]

[0045] sigma υ = 1

[0046] where Gamma(·) is the gamma function. ​

[0047] SS2. Calculate the fitness of each elephantfish.

[0048] SS3. Record the elephantfish with the lowest fitness as the optimal elephantfish and record its fitness value.

[0049] SS4. To more accurately detect the distribution of surrounding food, elephantfish can use radar to obtain information on the location and density of nearby food, thereby enhancing their local search capabilities. Radar detection is performed on all elephantfish to generate sampling points, specifically including:

[0050] For the i-th elephantfish, the number of sampling points S is generated according to the following formula. i and the detection area radius A i :

[0051]

[0052]

[0053] Among them, S max and S min These represent the maximum and minimum number of sampling points, respectively, and r1 and r2 are random numbers that follow a uniform distribution in the interval (0,1). and These represent the upper and lower limits of the value in the k-th dimension, respectively.

[0054] After obtaining the number of sampling points and the detection radius, the radar detection behavior of the elephantfish can be described as follows: A total of S... i Each sampling is performed using the following formula, with itself as the center and a radius of A. i Sampling point X is generated within the circular area. S :

[0055] X s =(X s,1 ,X s,2 ,...X s,D )

[0056] X s,k =A i +X k k = 1, 2, ..., D

[0057] Among them, X s,k Indicates sampling point X s The value of the k-th dimension, X k This represents the value of the k-th dimension of the elephantfish.

[0058] SS5. For each elephantfish, if radar detection reveals a sampling point with better fitness than the current location, a greedy selection strategy is adopted, and the elephantfish will move to that sampling point. The position update formula is as follows:

[0059] X t+1 = argmin(f(X i )), i = 1, 2,..., S i

[0060] where argmin(f(X i )) represents the value of X i that minimizes f(X i ).

[0061] S6, for each elephant fish, if a better sampling point is not found by radar detection, a resource competition strategy is adopted, then a random probability is selected to move to the optimal position of the population or a random position in the solution space, and the position updating formula is as follows:

[0062]

[0063] where X best represents the optimal elephant fish in the population, r3 is uniformly distributed in the interval (0, 1), p represents the probability of moving to the optimal elephant fish, X rand represents a random position in the solution space, calculated by the following formula:

[0064]

[0065] where r4 is uniformly distributed in the interval (0, 1).

[0066] S7, update the fitness of each elephant fish after moving the position.

[0067] S8, let t = t + 1, judge whether t = T is satisfied, if not, return to S3; if satisfied, execute step S10.

[0068] S9, consider that the operation is completed, output the optimal elephant fish in the last iteration, and the solution represented by the elephant fish is the generated optimal path.

[0069] S6, output the navigation instruction: convert the generated optimal path into navigation instruction to guide the robot to navigate in the environment.

[0070] Through the above implementation, the path planning method of the application can effectively utilize the environment map data and the elephant fish algorithm to realize efficient and robust path planning in autonomous navigation robots and other intelligent systems. Figure 3 For the two-dimensional map path planning result instance diagram of the application, compared with the ant colony algorithm and the differential evolution algorithm, the planned unmanned robot path is obviously better than the other two algorithms, and this method has a wide application prospect and can be used in the fields of autonomous driving vehicles, unmanned aerial vehicles and industrial robots.

[0071] The embodiments of the present application in conjunction with the drawings are described in detail above, but the present application is not limited to the described embodiments. For those skilled in the art, various changes, modifications, replacements and variations of the embodiments including components are made without departing from the principles and spirits of the present application, and still fall within the protection scope of the present application.

Claims

1. A path planning method based on elephant trunk optimization strategy, characterized in that, The method comprises the following steps: S1, starting the sensor system to scan the environment, generating an initial environment map, and generating a final environment map including the starting point, the ending point and the obstacle circle after preprocessing; S2, initializing the iteration number t=1, initializing N elephant trunks according to the Levy flight formula, one elephant trunk representing a path, and calculating the fitness of each elephant trunk; S3, saving the elephant trunk with the minimum fitness in the N elephant trunks as the optimal elephant trunk; S4, for each elephant trunk, performing radar detection to generate a sampling point, and calculating the fitness of the sampling point; The specific process of performing radar detection to generate a sampling point is as follows: For the first elephant-nose fish, the number of sampling points and the detection area radius are generated according to the following formula: ； ； wherein, and are the maximum and minimum of the number of sampling points, respectively, and are the upper and lower limits of the interval are random numbers subject to a uniform distribution, and denote the upper and lower limits of the value of the dimension, respectively. After the number of sampling points and the detection radius are obtained, the radar detection behavior of elephant nose fish is described as follows: a total of sampling points are generated, and each sampling point is generated in a circular region with a radius of around the elephant nose fish by using the following formula : ； ； wherein, represents the kth dimension value of the sampling point represents the kth dimension value of the sampling point represents the kth dimension value of the sampling point represents the kth dimension value of the sampling point represents the kth dimension value of the sampling point S5, for each elephant trunk, if a sampling point with better fitness than the current position is found, a greedy selection strategy is adopted, and the elephant trunk is moved to the sampling point; S6, for each elephant trunk, if no sampling point with better fitness is found, a resource contention strategy is adopted, and a random probability is selected to move to the optimal position of the population or a random position in the solution space; S7, calculating the fitness of each elephant trunk after updating; S8, t=t+1, judging whether t reaches the maximum iteration number, if not, returning to S2 to perform the next iteration operation, if yes, taking the final optimal elephant trunk as the global optimal path; S9, converting the generated path into navigation instructions to guide the robot to navigate in the environment.

2. The path planning method based on elephant trunk optimization strategy according to claim 1, characterized in that, The preprocessing in S1 is to perform the outer circle inflation on all obstacles in the initial environment map.

3. The path planning method based on the elephant trunk optimization strategy according to claim 1, characterized in that, The process of initializing the elephant trunk in S2 is as follows: ； wherein, is the newly generated elephant-nose fish, is the generated elephant-nose fish, the first elephant-nose fish of the population is randomly generated in the solution space, is the step size scaling factor, is the step size that is subject to a Levy distribution and is calculated according to the following equation: ； wherein , , take 1.5, and: ； ； wherein is the gamma function.

4. The path planning method based on the elephant trunk optimization strategy according to claim 3, characterized in that, The specific process of calculating the fitness of each elephant trunk in S2 is as follows: Path length: ； wherein, representing a key point and a path segment length of the key point ; Obstacle cost: ； wherein, is the cost coefficient of the obstacle, p(r) is the obstacle cost of the path passing through the key point A, R is the circumscribed circle radius of the obstacle, and r is the distance of the key point A from the center of the circle. Fitness function: ； where o is the number of obstacles, m and n are the number of key points and path sampling points, respectively, represents the distance from the first key point or sampling point of the path to the center of the circumscribed circle of the first obstacle.

5. The path planning method based on the elephant trunk optimization strategy according to claim 4, characterized in that, The specific process of selecting to move to the optimal position of the population or a random position in the solution space according to a random probability in S6 is as follows: The position updating formula is as follows: ； where, represents the best elephant in the population, is uniformly distributed in the interval represents the probability of moving towards the best elephant, represents a random position within the solution space, calculated by​ ； wherein In the interval obeys a uniform distribution.

Images