A method for gastrointestinal path planning in capsule robots based on adaptive mutated jackal algorithm

Abstract

This invention discloses a gastrointestinal path planning method for capsule robots based on an adaptive mutated Golden Jackal algorithm. The method includes image preprocessing and SIFT feature point extraction based on the complex and variable surface environment of the gastrointestinal tract; employing a fast nearest neighbor search algorithm based on local feature topological constraints for precise matching of image feature points; using the LM algorithm to reduce projection errors on the image imaging plane; and utilizing the Poisson surface reconstruction method to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract to obtain the reconstructed gastrointestinal condition. An adaptive mutated Golden Jackal algorithm is then introduced. Furthermore, Henon chaotic mapping theory, a nonlinear composite adaptive convergence factor random selection strategy, and a genetic algorithm mutation strategy are incorporated to improve the convergence accuracy and speed of the Golden Jackal algorithm, avoid the algorithm getting trapped in local optima, and achieve optimized path planning for the capsule robot. This addresses the complex environment of the gastrointestinal tract and provides assurance for high-precision control and gastrointestinal inspection of the capsule robot.

Description

Technical Field

[0001] This invention relates to a gastrointestinal pathway planning method, specifically a capsule robot gastrointestinal pathway planning method based on an adaptive mutant jackal algorithm, belonging to the field of pathway optimization technology for novel medical devices. Background Technology

[0002] In recent years, with the fast pace of life and increasing work pressure, most people's gastrointestinal tracts are in a sub-healthy state. With the continuous introduction and improvement of national policies, new medical devices such as capsule robots are constantly developing and being widely used in gastrointestinal examinations, effectively reducing pain and discomfort throughout the examination process.

[0003] However, the working environment of capsule robots is extremely complex. The rugged and variable working environment created by factors such as mucus and folds on the surface of the gastrointestinal tract poses great challenges to the active control and full-range inspection of capsule robots. Therefore, it is of great necessity to find an optimized path that meets the requirements of capsule robots in the complex environment of the gastrointestinal tract.

[0004] In recent years, intelligent optimization algorithms have been widely applied to path planning problems. Among them, Nitish Chopra and Muhammad Mohsin Ansari proposed the Golden Jackal Optimization (GJO) algorithm in 2022, which simulates the hunting behavior of golden jackals. This algorithm is novel, fast, and efficient. However, the algorithm itself is prone to getting trapped in local optima, and there are still some issues regarding convergence accuracy and convergence speed. Summary of the Invention

[0005] The purpose of this invention is to provide a gastrointestinal path planning method for capsule robots based on the adaptive mutated jackal algorithm in order to solve at least one of the above-mentioned technical problems.

[0006] This invention achieves the above objective through the following technical solution: a gastrointestinal path planning method for capsule robots based on an adaptive mutated jackal algorithm, comprising the following steps:

[0007] Step 1: Due to the complex and variable working environment of the stomach where the capsule robot is located, the images captured by the capsule robot's camera module are subject to complex noise interference, such as salt and pepper noise, uneven lighting, and gastric mucus. Therefore, the acquired images are first preprocessed, including bilateral filtering and image enhancement.

[0008] Step 2: Establish the scale space of the image and use the SIFT algorithm to extract feature points from the image;

[0009] Step 3: Search for feature points and use a fast nearest neighbor search algorithm based on local feature topological constraints to perform fine matching of image feature points;

[0010] Step 4: As the image matching points and cumulative projection errors accumulate during the reconstruction process, the LM algorithm is used to reduce the projection error of the image imaging plane.

[0011] Step 5: Use the Poisson surface reconstruction method to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract, and obtain the reconstructed gastrointestinal tract condition.

[0012] Step 6: Introduce the adaptive mutated golden jackal algorithm;

[0013] Step 7: Based on the obtained gastrointestinal information, pre-plan the operating path of the capsule robot.

[0014] As a further aspect of the present invention: In step six, the basic algorithm process of the Golden Jackal algorithm is as follows:

[0015] The hunting process of their entire population can be divided into three basic stages:

[0016] (1) Search for prey and approach it;

[0017] (2) Surround the prey and stimulate it until it stops moving;

[0018] (3) Attacking prey;

[0019] J0 = J min +rand×(J max -J min (1)

[0020] In the formula: J0 is the initial position of the golden jackal population; rand is a random number in the range [0,1]; J max and J min These are the upper and lower boundaries of the problem to be solved;

[0021] In the GJO algorithm, the prey matrix is ​​represented as:

[0022]

[0023] In the formula: Prey is the prey matrix; J i,j Let be the j-th position of the i-th prey; n be the size of the golden jackal population; and d be the solution dimension of the problem.

[0024] As a further aspect of this invention: in the adaptive mutation optimization process of the golden jackal algorithm, the fitness value of each golden jackal during the hunting process is calculated using a target fitness function, and its fitness value matrix is ​​represented as follows:

[0025]

[0026] In the formula: F PP is the fitness value matrix of the prey; f(...) is the fitness function or objective function; where the prey with the best fitness value is the male golden jackal, and the prey with the second best fitness value is the female golden jackal.

[0027] Golden jackal populations rely on their own senses to search for prey, and the amount of energy the prey uses to escape directly affects the population's behavior; E represents the prey's escape energy, which can be calculated using the following formula:

[0028] E = E0 × E1 (4)

[0029] Where E0 = 2 × rand-1 represents the initial state of the prey's energy, and E1 represents the process of the prey's energy decreasing;

[0030]

[0031] In the formula: T is the maximum number of iterations; c1 is a constant with a value of 1.5; t is the current iteration number; throughout the entire iteration process, E1 decreases linearly from 1.5 to 0;

[0032] In the above formula, rl is a random number vector based on the Levy distribution, expressed as:

[0033] rl=0.05×levy(j) (6)

[0034] Where levy() represents the Lévy flight function, which is calculated as follows:

[0035]

[0036]

[0037] In the formula: μ and v are random numbers in the range (0,1); β is a default constant with a value of 1.5;

[0038] When |E|≥1, the golden jackal population enters the search and exploration phase, with male jackals leading the hunting and female jackals following behind. Their group behavior is as follows:

[0039] J1(t)=J M (t)-E×|J M (t)-rl×Prey(t)| (9)

[0040] J2(t)=J FM (t)-E×|J FM (t)-rl×Prey(t)| (10)

[0041] In the formula: t is the current iteration number; Prey(t) is the position of the prey in the t-th iteration; J M (t),J FM J1(t) and J2(t) represent the positions of the male and female golden jackals in the t-th iteration, respectively; J1(t) and J2(t) represent the positions of the male and female golden jackals after the update in the t-th iteration, respectively; thus, the updated positions of the golden jackals after the (t+1)-th iteration are as follows:

[0042]

[0043] When |E| < 1, the golden jackal population enters the encirclement and attack phase, where the jackals surround and prey on the prey they found in the previous phase. The behavioral models of male and female golden jackals in this phase are as follows:

[0044] J1(t)=J M (t)-E×|rl×J M (t)-Prey(t)| (12)

[0045] J2(t)=J FM (t)-E×|rl×J FM (t)-Prey(t)| (13)

[0046] The positions of male and female golden jackals are updated using the above method after the t-th iteration, and the position update of the golden jackals is still completed according to formula (11).

[0047] As a further aspect of the present invention: based on the golden jackal optimization algorithm, Henon chaotic mapping theory is introduced for the initialization stage of the golden jackal population.

[0048] Henon's chaotic mapping theory is a nonlinear theory that arises in two-dimensional space. It is a typical discrete chaotic mapping, and its dynamic formula is as follows:

[0049]

[0050] The four parameters x0, y0, a, and b determine the state of the Henon chaotic map, which is more complex than the one-dimensional chaotic map. When a = 1.3 and b = 0.4, the strong randomness of the generated chaotic sequence is guaranteed when the function enters the chaotic state, thus obtaining the chaotic initialization population.

[0051] As a further aspect of the present invention: based on the golden jackal optimization algorithm, a nonlinear composite adaptive convergence factor random selection strategy is introduced for the search and predation phases of the golden jackal population.

[0052] First, with gama set to 0.9 and a starting threshold yz = rand introduced, the formula for selecting the convergence factor is as follows:

[0053]

[0054] And this convergence factor is then incorporated into the search and predation phases, and the position update formula for the search phase is modified as follows:

[0055] J1(t)=ω×J M (t)-E×|J M (t)-rl×Prey(t)| (16)

[0056] J2(t)=ω×J FM (t)-E×|J FM (t)-rl×Prey(t)| (17)

[0057] The position update formula during the predation phase is as follows:

[0058] J1(t)=ω×J M (t)-E×|rl×J M (t)-Prey(t)| (18)

[0059] J2(t)=ω×J FM (t)-E×|rl×J FM (t)-Prey(t)| (19)

[0060] The introduction of this convergence factor can greatly increase the convergence speed of the Golden Cow algorithm, significantly improve the optimization accuracy, and supplement the algorithm's need for global search capability and local search capability at different stages.

[0061] As a further aspect of the present invention: based on the golden jackal optimization algorithm, a genetic algorithm mutation strategy is introduced for the output stage of the golden jackal population.

[0062] The mutation probability Pm = 0.2 is given in advance.

[0063]

[0064] This method can effectively increase the diversity of solutions, thereby increasing the possibility of escaping local extrema and making the overall optimization effect of the improved algorithm better.

[0065] The beneficial effects of this invention are as follows: Based on the complex and variable surface environment of the gastrointestinal tract, image preprocessing and SIFT feature point extraction are first performed. A fast nearest neighbor search algorithm based on local feature topological constraints is then used to precisely match the image feature points. The LM algorithm is employed to reduce the projection error of the image imaging plane. Finally, the Poisson surface reconstruction method is used to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract, obtaining the reconstructed gastrointestinal condition. For the gastrointestinal condition obtained from the three-dimensional reconstruction, an adaptive mutation Golden Jackal algorithm is introduced. This algorithm, based on the original algorithm, incorporates Henon chaotic mapping theory, a nonlinear composite adaptive convergence factor random selection strategy, and a genetic algorithm mutation strategy. Through these improvements, the diversity of the population is increased, the global search capability and local search capability of the algorithm are balanced, and the convergence accuracy and convergence speed of the Golden Jackal algorithm are effectively improved, avoiding the algorithm from getting trapped in local optima. The improved Golden Jackal algorithm enables optimized path planning for capsule robots, providing a guarantee for high-precision control of capsule robots and gastrointestinal inspection. Attached Figure Description

[0066] Figure 1 This is a simplified flowchart of the present invention.

[0067] Figure 2 This is a flowchart of the adaptive mutated golden jackal algorithm in this invention.

[0068] Figure 3 This is a convergence curve of the first test function in this embodiment of the invention under the adaptive mutated jackal algorithm and several other intelligent optimization algorithms.

[0069] Figure 4 This is a convergence curve of the second test function in this embodiment of the invention under the adaptive mutated jackal algorithm and several other intelligent optimization algorithms.

[0070] Figure 5 This is a convergence curve of the third test function in this embodiment of the invention under the adaptive mutated jackal algorithm and several other intelligent optimization algorithms.

[0071] Figure 6 This refers to a commonly used simulation environment map in the embodiments of the present invention;

[0072] Figure 7 This is a simulation result diagram of a commonly used simulation environment map in the embodiments of the present invention. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0074] Example 1

[0075] like Figures 1 to 7 As shown, a gastrointestinal pathway planning method for capsule robots based on the adaptive mutated jackal algorithm includes the following steps:

[0076] Step 1: Due to the complex and variable working environment of the stomach where the capsule robot is located, the images captured by the capsule robot's camera module are subject to complex noise interference, such as salt and pepper noise, uneven lighting, and gastric mucus. Therefore, the acquired images are first preprocessed, including bilateral filtering and image enhancement.

[0077] Step 2: Establish the scale space of the image and use the SIFT algorithm to extract feature points from the image;

[0078] Step 3: Search for feature points and use a fast nearest neighbor search algorithm based on local feature topological constraints to perform fine matching of image feature points;

[0079] Step 4: As the image matching points and cumulative projection errors accumulate during the reconstruction process, the LM algorithm is used to reduce the projection error of the image imaging plane.

[0080] Step 5: Use the Poisson surface reconstruction method to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract, and obtain the reconstructed gastrointestinal tract condition.

[0081] Step 6: Introduce the adaptive mutated golden jackal algorithm;

[0082] Step 7: Based on the obtained gastrointestinal information, pre-plan the operating path of the capsule robot.

[0083] Example 2

[0084] In addition to all the technical features included in Embodiment 1, this embodiment also includes:

[0085] In step six, the basic algorithm process of the Golden Jackal algorithm is as follows:

[0086] The hunting process of their entire population can be divided into three basic stages:

[0087] (1) Search for prey and approach it;

[0088] (2) Surround the prey and stimulate it until it stops moving;

[0089] (3) Attacking prey;

[0090] J0 = J min +rand×(J max -J min (1)

[0091] In the formula: J0 is the initial position of the golden jackal population; rand is a random number in the range [0,1]; J max and J min These are the upper and lower boundaries of the problem to be solved;

[0092] In the GJO algorithm, the prey matrix is ​​represented as:

[0093]

[0094] In the formula: Prey is the prey matrix; J i,j Let be the j-th position of the i-th prey; n be the size of the golden jackal population; and d be the solution dimension of the problem.

[0095] In the adaptive mutation optimization process of the Golden Jackal algorithm, the fitness value of each Golden Jackal during the hunting process is calculated using the target fitness function. The fitness value matrix is ​​represented as follows:

[0096]

[0097] In the formula: F PP is the fitness value matrix of the prey; f(...) is the fitness function or objective function; where the prey with the best fitness value is the male golden jackal, and the prey with the second best fitness value is the female golden jackal.

[0098] Golden jackal populations rely on their own senses to search for prey, and the amount of energy the prey uses to escape directly affects the population's behavior; E represents the prey's escape energy, which can be calculated using the following formula:

[0099] E = E0 × E1 (4)

[0100] Where E0 = 2 × rand-1 represents the initial state of the prey's energy, and E1 represents the process of the prey's energy decreasing;

[0101]

[0102] In the formula: T is the maximum number of iterations; c1 is a constant with a value of 1.5; t is the current iteration number; throughout the entire iteration process, E1 decreases linearly from 1.5 to 0;

[0103] In the above formula, rl is a random number vector based on the Levy distribution, expressed as:

[0104] rl=0.05×levy(j) (6)

[0105] Where levy() represents the Lévy flight function, which is calculated as follows:

[0106]

[0107]

[0108] In the formula: μ and v are random numbers in the range (0,1); β is a default constant with a value of 1.5;

[0109] When the calculated value of E ≥ 1, the golden jackal population enters the search and exploration phase, with male jackals leading the hunting and female jackals following behind. Their group behavior is as follows:

[0110] J1(t)=J M (t)-E×|J M (t)-rl×Prey(t)| (9)

[0111] J2(t)=J FM (t)-E×|J FM (t)-rl×Prey(t)| (10)

[0112] In the formula: t is the current iteration number; Prey(t) is the position of the prey in the t-th iteration; J M (t),J FM J1(t) and J2(t) represent the positions of the male and female golden jackals in the t-th iteration, respectively; J1(t) and J2(t) represent the positions of the male and female golden jackals after the update in the t-th iteration, respectively; thus, the updated positions of the golden jackals after the (t+1)-th iteration are as follows:

[0113]

[0114] When |E| < 1, the golden jackal population enters the encirclement and attack phase, where the jackals surround and prey on the prey they found in the previous phase. The behavioral models of male and female golden jackals in this phase are as follows:

[0115] J1(t)=J M (t)-E×|rl×J M (t)-Prey(t)| (12)

[0116] J2(t)=J FM (t)-E×|rl×J FM(t)-Prey(t)| (13)

[0117] The positions of male and female golden jackals are updated using the above method after the t-th iteration, and the position update of the golden jackals is still completed according to formula (11).

[0118] Example 3

[0119] In addition to all the technical features included in Embodiment 1, this embodiment also includes:

[0120] Based on the golden jackal optimization algorithm, Henon chaotic mapping theory is introduced for the initialization stage of the golden jackal population.

[0121] Henon's chaotic mapping theory is a nonlinear theory that arises in two-dimensional space. It is a typical discrete chaotic mapping, and its dynamic formula is as follows:

[0122]

[0123] The four parameters x0, y0, a, and b determine the state of the Henon chaotic map, which is more complex than the one-dimensional chaotic map. When a = 1.3 and b = 0.4, the strong randomness of the generated chaotic sequence is guaranteed when the function enters the chaotic state, thus obtaining the chaotic initialization population.

[0124] Based on the optimization algorithm for golden jackals, a nonlinear composite adaptive convergence factor random decision strategy is introduced for the search and predation phases of the golden jackal population.

[0125] First, with gama set to 0.9 and a starting threshold yz = rand introduced, the formula for selecting the convergence factor is as follows:

[0126]

[0127] And this convergence factor is then incorporated into the search and predation phases, and the position update formula for the search phase is modified as follows:

[0128] J1(t)=ω×J M (t)-E×|J M (t)-rl×Prey(t)| (16)

[0129] J2(t)=ω×J FM (t)-E×|J FM (t)-rl×Prey(t)| (17)

[0130] The position update formula during the predation phase is as follows:

[0131] J1(t)=ω×J M (t)-E×|rl×JM (t)-Prey(t)| (18)

[0132] J2(t)=ω×J FM (t)-E×|rl×J FM (t)-Prey(t)| (19)

[0133] The introduction of this convergence factor can greatly increase the convergence speed of the Golden Cow algorithm, significantly improve the optimization accuracy, and supplement the algorithm's need for global search capability and local search capability at different stages.

[0134] Based on the Golden Jackal optimization algorithm, a genetic algorithm mutation strategy is introduced for the output stage of the Golden Jackal population.

[0135] The mutation probability Pm = 0.2 is given in advance.

[0136]

[0137] This method can effectively increase the diversity of solutions, thereby increasing the possibility of escaping local extrema and making the overall optimization effect of the improved algorithm better.

[0138] Example 4

[0139] A method for gastrointestinal path planning in capsule robots based on an adaptive mutated jackal algorithm, employing a basic test function. To verify the performance of the improved algorithm, the theoretical optimal value of the function is 0. The improved Golden Jackal algorithm is compared with the basic Golden Jackal algorithm, the Gray Wolf optimization algorithm, the Northern Eagle optimization algorithm, the Hunger Games search algorithm, the mouse swarm optimization algorithm, and the Harris Eagle optimization algorithm to verify the performance of the improved Golden Jackal optimization algorithm. To ensure fairness in the testing, the population size for each algorithm is set to 30, and the maximum number of iterations is set to 500.

[0140] like Figure 3 As shown, the adaptive mutant jackal algorithm has significant advantages over other algorithms in terms of convergence speed and convergence accuracy.

[0141] Example 5

[0142] A method for gastrointestinal path planning in capsule robots based on an adaptive mutated jackal algorithm, employing a basic test function. To verify the performance of the improved algorithm, the theoretical optimal value of the function is 0. The improved Golden Jackal algorithm is compared with the basic Golden Jackal algorithm, the Gray Wolf optimization algorithm, the Northern Eagle optimization algorithm, the Hunger Games search algorithm, the mouse swarm optimization algorithm, and the Harris Eagle optimization algorithm to verify the performance of the improved Golden Jackal optimization algorithm. To ensure fairness in the testing, the population size for each algorithm is set to 30, and the maximum number of iterations is set to 500.

[0143] like Figure 4 As shown, the adaptive mutant jackal algorithm has significant advantages over other algorithms in terms of convergence speed and convergence accuracy.

[0144] Example 6

[0145] A method for gastrointestinal path planning in capsule robots based on an adaptive mutated jackal algorithm, employing a basic test function. To verify the performance of the improved algorithm, the theoretical optimal value of the function is 1. The improved Golden Jackal algorithm is compared with the basic Golden Jackal algorithm, the Gray Wolf optimization algorithm, the Northern Eagle optimization algorithm, the Hunger Games search algorithm, the mouse swarm optimization algorithm, and the Harris Eagle optimization algorithm to verify the performance of the improved Golden Jackal optimization algorithm. To ensure fairness in the testing, the population size for each algorithm is set to 30, and the maximum number of iterations is set to 500.

[0146] like Figure 5 As shown, the adaptive mutant jackal algorithm has significant advantages over other algorithms in terms of convergence speed and convergence accuracy.

[0147] Example 7

[0148] A gastrointestinal path planning method for capsule robots based on the adaptive mutant jackal algorithm is presented. To further verify the advantages and machine applicability of the proposed algorithm, this invention compares and analyzes the adaptive mutant jackal algorithm with the basic jackal algorithm in a 30*30 simulation environment map. Figure 6 The image shown is a 30x30 simulation environment map. The simulation results after the algorithm is run are as follows: Figure 7 As shown in the figure, the adaptive mutated jackal algorithm can obtain the optimal path length more quickly and complete path planning.

[0149] Working principle: Based on the complex and variable surface environment of the gastrointestinal tract, image preprocessing and SIFT feature point extraction are performed. A fast nearest neighbor search algorithm based on local feature topological constraints is used to perform fine matching of image feature points. The LM algorithm is used to reduce the projection error of the image imaging plane. Finally, the Poisson surface reconstruction method is used to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract, obtaining the reconstructed gastrointestinal tract condition.

[0150] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0151] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A gastrointestinal pathway planning method for capsule robots based on an adaptive mutated jackal algorithm, characterized in that, The capsule robot gastrointestinal pathway planning method includes the following steps: Step 1: Preprocess the acquired images; Step 2: Establish the scale space of the image and use the SIFT algorithm to extract feature points from the image; Step 3: Search for feature points and use a fast nearest neighbor search algorithm based on local feature topological constraints to perform fine matching of image feature points; Step 4: As the image matching points and cumulative projection errors accumulate during the reconstruction process, the LM algorithm is used to reduce the projection error of the image imaging plane. Step 5: Use the Poisson surface reconstruction method to perform three-dimensional reconstruction of the actual situation inside the gastrointestinal tract, and obtain the reconstructed gastrointestinal tract condition. Step 6: Introduce the adaptive mutated golden jackal algorithm; Step 7: Based on the obtained gastrointestinal information, pre-plan the operating path of the capsule robot; In step six, the basic algorithm process of the Golden Jackal algorithm is as follows: The hunting process of their entire population can be divided into three basic stages: (1) Search for prey and approach it; (2) Surround the prey and stimulate it until it stops moving; (3) Attacking prey; ； In the formula: This indicates the location of the initial golden jackal population. It is a random number in the range [0,1]. and These are the upper and lower boundaries of the problem to be solved; In the GJO algorithm, the prey matrix is ​​represented as: ； In the formula: For the prey matrix; For the first The first prey Dimensional position; This refers to the population size of the golden jackal; The dimensions for solving the problem; In the adaptive mutation optimization process of the Golden Jackal algorithm, the fitness value of each Golden Jackal during the hunting process is calculated using the target fitness function, and its fitness value matrix is ​​represented as follows: ； In the formula: This is the fitness value matrix of the prey; The fitness function or objective function; Among them, the one with the best fitness value is designated as the male golden jackal, and the one with the second best fitness value is designated as the female golden jackal. Golden jackal populations rely on their own senses to search for prey, and the amount of energy the prey uses to escape directly affects the population behavior of golden jackals. The escape energy of the prey is calculated using the following formula: ； in, This indicates the initial energy state of the prey. This indicates the process of the prey's energy decreasing; ； In the formula: This represents the maximum number of iterations. It is a constant with a value of 1.5; This represents the current iteration number; throughout the entire iteration process, It decreases linearly from 1.5 to 0; The formula mentioned above It is based on A vector of randomly distributed numbers is represented as: ； in The Lévy flight function is represented and its calculation method is as follows: ； ； In the formula: and A random number within the range (0,1); To use the default constant, the value is 1.5; When the desired result is obtained At this time, the golden jackal population enters the search and exploration phase, with male jackals leading the hunting and female jackals following behind. Their group behavior is as follows: ； ； In the formula: This represents the current iteration number; For the first The location of the prey in the next iteration; , The first The positions of the male and female golden jackals in the next iteration; , The first The locations of the male and female golden jackals after the next iteration update; Thus, the first The updated positions of the golden jackals after the next iteration are as follows: ； When the desired result is obtained At this time, the golden jackal population enters the encirclement and attack phase, where the golden jackals surround and prey on the prey they have searched for in the previous phase. The behavioral patterns of male and female golden jackals at this stage are as follows: ； ； The above method is used to update and obtain the first... The positions of male and female golden jackals after each iteration are determined, and the positions of the golden jackals are updated according to formula (11).

2. The capsule robot gastrointestinal pathway planning method according to claim 1, characterized in that: Based on the golden jackal optimization algorithm, Henon chaotic mapping theory is introduced for the initialization stage of the golden jackal population. Henon's chaotic mapping theory is a nonlinear theory that arises in two-dimensional space. It is a typical discrete chaotic mapping, and its dynamic formula is as follows: ； in, , , and These four parameters determine the state of the Henon chaotic map, which is more complex than the one-dimensional chaotic map. When a=1.3 and b=0.4, the strong randomness of the generated chaotic sequence is guaranteed when the function enters the chaotic state, thus obtaining the chaotic initialization population.

3. The capsule robot gastrointestinal pathway planning method according to claim 1, characterized in that: Based on the optimization algorithm for golden jackals, a nonlinear composite adaptive convergence factor random decision strategy is introduced for the search and predation phases of the golden jackal population. First, with gama set to 0.9 and a starting threshold yz = rand introduced, the formula for selecting the convergence factor is as follows: ； And this convergence factor is then incorporated into the search and predation phases, and the position update formula for the search phase is modified as follows: ； ； The position update formula during the predation phase is as follows: ； ； The introduction of this convergence factor can greatly increase the convergence speed of the Golden Cow algorithm, significantly improve the optimization accuracy, and supplement the algorithm's need for global search capability and local search capability at different stages.

4. The capsule robot gastrointestinal pathway planning method according to claim 1, characterized in that: Based on the golden jackal optimization algorithm, a genetic algorithm mutation strategy is introduced for the output stage of the golden jackal population: Given a mutation probability Pm = 0.2, ； This method increases the diversity of solutions, thereby increasing the likelihood of escaping local optima and making the overall optimization performance of the improved algorithm better.

Images