CN116807611A - Flexible puncture needle path planning method based on differential evolution algorithm - Google Patents

Abstract

The application relates to the technical field of intelligent medical auxiliary robot control, in particular to a flexible puncture needle path planning method based on a differential evolution algorithm, which comprises the following steps: acquiring environmental information of a focus part through an image collector; establishing a flexible needle puncture model and a path optimization objective function; determining control parameters of a differential evolution algorithm, randomly generating an initial population, and evaluating the initial population; selecting individuals from the initial population according to the fitness value to perform mutation operation to obtain variant individuals; performing cross operation on the variant individuals and target individuals to obtain experimental individuals; in the evolution process, carrying out boundary condition processing operation to obtain a temporary population; and selecting optimal individuals from the individuals in the temporary population and the target individuals by using a greedy strategy to form the target population. The algorithm introduced by the application has superior performance, better reliability, high efficiency and robustness, and stronger global convergence capability.

Description

A flexible puncture needle path planning method based on differential evolution algorithm

Technical field

The invention relates to the technical field of intelligent medical auxiliary robot control, specifically a flexible puncture needle path planning method based on a differential evolution algorithm.

Background technique

Minimally invasive surgical technology has the advantages of less trauma, less bleeding, and short postoperative recovery time, and is an important development direction in clinical surgery. In minimally invasive percutaneous puncture treatment, in order to reduce secondary damage to the patient, doctors need to bypass obstacles such as large blood vessels or nerves to reach the affected area. However, the steel needles currently used can only move in a straight line and cannot achieve deep penetration of the human body. Targeted puncture. As for flexible needles, the stiffness of the needle shaft is small, and the force on the needle tip is unbalanced, causing the needle shaft to deflect and produce a curved trajectory to avoid nerves, blood vessels and other important organs and bones and other obstacles, making it more flexible and precise to reach areas that traditional rigid needles cannot reach. target location. Therefore, the study of path planning of flexible needles is of great significance for accurate puncture, and the puncture motion of flexible needles requires precise planning and optimization.

Reference [1] divides flexible needle path planning into three categories according to algorithms: numerical method [2-4], search method [5-8] and inverse solution method [9]. The numerical method includes probability method and objective function method. The probability method considers the randomness of path errors and uses mathematical probability principles to calculate the path with the highest puncture success rate; the objective function method combines some optimization parameters (such as the shortest path, the shortest distance to obstacles, etc.) to establish an objective function and calculate the target The function obtains the optimal solution; although the numerical method is accurate in calculation, it is sometimes unable to adapt in complex environments (multiple irregular obstacles). The function of the search method is to search for feasible paths, but it does not guarantee that it is the optimal path; the inverse solution method does not guarantee that there is a solution, let alone the optimal solution. It can be seen that the above algorithm still needs to be improved for the path planning of flexible needles. In the flexible needle puncture path planning, the randomness and complexity of the path are high, the convergence speed is slow, and it is easy to fall into the local optimal solution. However, we apply the differential evolution algorithm to the flexible needle path planning and use its strong Intrinsic parallelism improves convergence speed.

The differential evolution algorithm is an optimization algorithm based on the group intelligence theory. It is an intelligent optimization search algorithm generated through cooperation and competition among individuals within the group. But compared with evolutionary calculation, it reduces the complexity of evolutionary calculation operations. At the same time, it has strong global convergence ability and robustness, and does not require the use of characteristic information of the problem. It is suitable for solving some complex optimization problems that are difficult or even impossible to solve using conventional mathematical programming methods [10-12]. Therefore, the differential evolution algorithm is an efficient parallel search algorithm, and its theoretical and applied research has important academic significance and engineering value. However, when solving the problem of flexible needle movement path planning for puncture robots, due to the particularity of flexible needle movement, the planned path must conform to its own movement rules. Therefore, it is necessary to establish a path planning scenario and puncture path model for puncture surgery, and at the same time establish The objective function is to achieve the purpose of optimal path planning and practical application in this scenario.

The differential evolution algorithm introduced in this invention is an efficient and powerful global optimization algorithm that guides the direction of optimization search through the group intelligence generated by mutual cooperation and competition among individuals in the group. Compared with the algorithms introduced above, this algorithm mainly has three major advantages: superior performance, good reliability, efficiency and robustness; adaptability, the differential mutation operator of this algorithm can be performed according to different objective functions Automatic adjustment improves the search quality; inherent parallelism, the ability to use individual local information and group global information to guide the algorithm for further search; low parameter sensitivity. Under the same accuracy requirements, the differential evolution algorithm has a faster convergence speed.

References are as follows:

[1] Zhang Yongde. Review of flexible needle puncture path planning [J]. Journal of Harbin University of Science and Technology, 2011, (4): 7-11.

[2] Lee J, Park W. Insertion planning for steerable flexible needlesreaching multiple planar targets. New Horizon. 2015; 40: 2377–2383.

[3] Sun W, Alterovitz R. Motion planning under uncertainty for medical needle steering using optimization in belief space. IEEE International Conference on Robotics and Automation; 2014; Hongkong, China.

[4]Huo B, Zhao –202.

[5]Alterovitz R, Patil S, Derbakova A. Rapidly-exploring roadmaps: weighting exploration vs.refinement in optimal motion planning. Proceedingsunder IEEE International Conference on Robotics and Automation; 2011; Shanghai, China.p.3706–3716.

[6] Caborni C, Ko S, De M, et al. Risk-based path planning for asteerable flexible probe for neurosurgical intervention. IEEE RAS&EMBS International Conference on Biomedical Robotics and Biomechatronics; 2012; Rome, Italy.p.866–871.

[7] Bernardes M, Adorno B, Poignet P, et al. Robotassisted automatic insertion of steerable needles with Closed-Loop imaging feedback and intraoperative trajectory replanning. Mechatronics. 2013; 23(6): 630–645.

[8] Li X, Li P, Xiong J. Path planning for flexible needle based onenvironment properties and random method. Comput Eng Appl. 2017; 53:121–125.

[9]Duindam V, Xu J, Alterovitz R, et al. 3D motion planning algorithms for steerable needles using inverse kinematics. 8th International Workshop on the Algorithmic Foundations of Robotics; 2008; Guanajuato,Mexico.p.535–549.

[10]Storn R,Price K.Differential evolution-a simple and efficientheuristic for global optimization over continuous spaces[J].Journal of globaloptimization,1997,11(4):341.

[11]Kurup D G,Himdi M,Rydberg A.Synthesis of uniform amplitudeunequally spaced antenna arrays using the differential evolution algorithm[J].IEEE Transactions on Antennas and Propagation,2003,51(9):2210-2217.

[12] Bao Ziyang, Chen Kesong, He Zishu, Han Chunlin. Circular array sparse method based on improved differential evolution algorithm [J]. Systems Engineering and Electronic Technology, 2009, 31(03): 497-499.

Contents of the invention

In order to solve the above technical problems, the present invention proposes a flexible puncture needle path planning method based on a differential evolution algorithm.

The technical problems to be solved by the present invention are achieved by adopting the following technical solutions:

A flexible puncture needle path planning method based on differential evolution algorithm, including the following steps:

Step (1) Obtain the puncture environment image through the image collector, and determine the location of the target and obstacles from the image;

Step (2) Establish the flexible needle puncture model and path optimization objective function;

Step (3) Determine the control parameters of the differential evolution algorithm, randomly generate an initial population, and evaluate the initial population, that is, calculate the fitness value corresponding to each individual in the initial population;

Step (4) Select individuals from the initial population according to the fitness value to perform mutation operations to obtain mutated individuals;

Step (5) Perform a crossover operation on the mutant individual and the target individual to obtain an experimental individual;

Step (6) During the evolution process, perform boundary condition processing operations to obtain a temporary population;

Step (7) Use a greedy strategy for the individuals in the temporary population and the target individual to select the optimal individual to form the target population;

Step (8) determines whether the termination condition is met or the maximum number of evolutions is reached: if so, the evolution is terminated and the best individual at this time is output as the solution; otherwise, steps (4) to (7) are repeated until the conditions are met.

Preferably, the flexible needle puncture model Path established in step (2) is as follows:

In the formula, Path is composed of n segments, p _i represents the starting point of the i-th path, θ _i represents the central angle of the i-th arc, r _i represents the radius of the i-th arc, if the path is a straight line, then r _i The value is 0, ω _i represents the arc tangent vector at the i-th path segment, and _Li represents the path function of the i-th segment.

Preferably, the path optimization objective function Y in step (2) is specifically as follows:

Y＝μ ₁ Y _n +μ ₂ Y _d +μ ₃ Y _l

In the formula, Y _n represents the limiting factor as a function of the error between the final puncture position and the target position, Y _d represents the limiting factor as a function of the distance between the trajectory and the obstacle, Y _l represents the limiting factor as a function of the path length, μ ₁ , μ ₂ , and μ ₃ are the weighting coefficients of Y _n , Y _d , and Y _l respectively, where μ ₁ > μ ₂ > μ ₃ shows the importance of the three factors.

Preferably, the various relationships are as follows:

In the formula, l represents the allowed working length of the flexible needle; n represents the number of segments of the planned path; w represents the deviation between the final puncture position and the target position; w _max is the maximum allowed deviation between the final puncture position and the target position; l _i is the length of the path segment in the planned path; d is the minimum distance from obstacles in the planned path; d _min is the minimum allowed safe distance from obstacles.

Preferably, in step (3), the control parameters of the differential evolution algorithm are determined and the initial population is randomly generated using the following formula:

In the _{formula ,} and/> represents the limit of the parameter variable, rand(0,1) represents a random number in the interval [0,1], and NP represents the population size.

Preferably, the mutation operation formula in step (4) is as follows:

v _i (g)＝x _r1 (g)+F·(x _r2 (g)-x _r3 (g))

i≠r1≠r2≠r3

In the formula, x _r1 (g), x _r2 (g), x _r3 (g) represent three randomly selected individuals that are different from each other from the current group, and should not be the same as the target individual x _i , v _i (g ) represents the mutated individual corresponding to the target individual x _i (g), and F represents the scaling factor.

Preferably, the crossover operation formula in step (5) is as follows:

In the formula, rand(0,1) represents a random number in the interval [0,1], j _rand represents an integer randomly selected from 1 to D, and CR represents the crossover operator.

Preferably, in step (6) during the evolution process, it is judged whether each component in the mutant individual meets the boundary conditions. If the boundary conditions are not met, the mutant individual is regenerated using the generation method in step (3).

Preferably, the greedy strategy formula in step (7) is as follows:

The beneficial effects of the present invention are:

The present invention realizes flexible needle puncture path planning based on a differential evolution algorithm. Because it has inherent parallelism and can be searched collaboratively, it has the ability to use individual local information and group global information to guide the algorithm for further search. Under the same accuracy requirements, the differential evolution algorithm has faster convergence speed, better reliability, efficiency and robustness. It can provide a reasonable path plan for flexible needle puncture and reduce the complexity of evolutionary calculation operations.

Description of the drawings

The present invention will be further described below in conjunction with the accompanying drawings and examples:

Figure 1 is a flow chart of the present invention;

Figure 2 is a path planning diagram of the flexible needle puncture based on the differential evolution algorithm provided by the present invention in the absence of obstacles;

Figure 3 is a path planning diagram of the flexible needle puncture provided by the present invention based on the differential evolution algorithm when there are obstacles;

Figure 4 is a comparison chart of algorithm iteration curves for flexible needle puncture path planning based on differential evolution algorithms provided by the present invention.

Detailed ways

It should be understood that the specific embodiments described here are only used to explain the present invention and are not intended to limit the present invention.

Minimally invasive interventional medicine is a brand-new medical technology that has developed rapidly in recent years. It is an emerging medical method between surgery and internal medicine. Minimally invasive interventional therapy is performed under image guidance. It uses minimal trauma, no skin incision, and only puncture needle holes to insert instruments or drugs into the diseased tissue to perform physical, mechanical or chemical treatment. Among them, targeted puncture technology is one of the most promising medical methods. However, traditional puncture surgeries currently use rigid needles, which cannot effectively control the needle tip position. During puncture, the needle axis will bend slightly due to uneven force on the needle tip and tissue deformation, causing the needle tip to deviate from the intended target, leading to puncture failure. In response to the shortcomings of traditional puncture needles, Johns Hopkins University in the United States proposed the use of flexible needle control technology in 2004. Compared with the rigid needles used in traditional puncture surgeries, flexible needles are flexible enough during use and can bend the needle axis. , the curved trajectory of the needle shaft can be used to avoid obstacles such as nerves, blood vessels and other important organs and bones, and more flexibly and accurately reach the target position that traditional rigid needles cannot reach. In clinical application, it can reduce tissue incisions, reduce patient discomfort, and speed up postoperative recovery. Due to the movement characteristics of the flexible needle itself, as well as the presence of sensitive tissues (such as nerves and blood) and obstacles (such as bones), the puncture trajectory of the flexible needle in the body requires high precision, which requires high precision control of the flexible needle. Flexibility Needle path planning is the basis for robot-assisted flexible needle puncture control. Therefore, it is necessary to use visual images to reasonably plan a movement path of a flexible puncture needle based on the needle entry point and target target position. This is important for the precise control of the robot-assisted puncture system. significance.

Flexible needle path planning, as a key technology in the minimally invasive field, is one of the main research contents of flexible needle motion planning. Movement rules consist of path planning and trajectory planning. The sequence of points or curves connecting the starting point and the end point is called a path, and the strategy that constitutes the path is called path planning. Many scholars at home and abroad have conducted research on flexible needle path planning and achieved many results. Since the early 1960s, many swarm optimization algorithms have been introduced, ranging from ant colony algorithms to differential evolution algorithms. All these algorithms show their potential to solve many optimization problems. Swarm Intelligence (SI) has aroused the interest of many researchers in various fields. Bonabeau defines SI as "the emergent collective intelligence of a population of simple agents." SI is the collective intelligent behavior of self-organizing and decentralized systems, for example, artificial groups of simple agents. Examples include group foraging, cooperative transportation, nesting, collective sorting and clustering of social insects.

There are many types of swarm intelligence optimization algorithms, among which the representative ones include ant colony algorithm, particle swarm algorithm, genetic algorithm and differential evolution algorithm. The ant colony algorithm itself is derived from a natural phenomenon of biology, that is, the behavior of ants searching for food and finding paths. It is a simulated evolutionary algorithm. Diversity and positive feedback. However, if there is excessive diversity when this algorithm is applied to path planning, the entire system will be too active, resulting in too many random movements. When the diversity is insufficient and the positive feedback is too strong, rigidity will occur and the system cannot be adjusted in time. Particle swarm algorithm is a swarm intelligence algorithm designed by simulating the predatory behavior of a flock of birds. Starting from a random solution, iteratively finds the optimal solution, and evaluates the quality of the solution through fitness. It has the characteristics of ease, high accuracy, and fast convergence, but it is easy to fall into the local optimal solution. Genetic algorithm is a method of searching for optimal solutions by simulating the natural evolution process. It draws on Darwin's theory of biological evolution and Mendel's laws of inheritance to simulate the process of natural selection. In each generation of the genetic algorithm, processes similar to DNA mutation, crossover, replication, inheritance, etc. in the theory of evolution are carried out based on the fitness value of the individual. The algorithm has strong global search capabilities, but weak local search capabilities, and is prone to local convergence, Problems such as premature maturity, low efficiency, and poor stability.

The differential evolution algorithm introduced in this invention is an efficient and powerful global optimization algorithm that guides the direction of optimization search through the group intelligence generated by mutual cooperation and competition among individuals in the group. The basic idea is: starting from a randomly generated initial population, a new individual is generated by summing the vector difference of any two individuals in the population with a third individual, and then comparing the new individual with the corresponding individual in the contemporary population. , if the fitness of the new individual is better than the fitness of the current individual, the new individual will replace the old individual in the next generation, otherwise the old individual will be retained. Through continuous evolution, excellent individuals are retained and inferior individuals are eliminated, guiding the search to approach the optimal solution. Compared with the algorithms introduced above, this algorithm mainly has three major advantages: superior performance, good reliability, efficiency and robustness; adaptability, the differential mutation operator of this algorithm can be performed according to different objective functions Automatic adjustment improves search quality; inherent parallelism has the ability to use individual local information and group global information to guide the algorithm for further search. Under the same accuracy requirements, the differential evolution algorithm has a faster convergence speed.

This application is to solve the problem of flexible needle puncture path in the existing technology. Due to the high randomness and complexity of the path in the flexible needle puncture path planning, the convergence speed is slow, and it is easy to fall into the local optimal solution. We will The differential evolution algorithm is applied to the path planning of flexible needles, using its strong internal parallelism to improve the convergence speed.

In order to make it easy to understand the technical means, creative features, objectives and effects of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and examples.

As shown in Figures 1 to 4, a flexible puncture needle path planning method based on differential evolution algorithm includes the following steps:

Step (1) Obtain the puncture environment image through the image collector, and determine the location of the target and obstacles from the image.

Step (2) Establish the flexible needle puncture model and path optimization objective function.

Establishing a flexible needle puncture model In practical applications, there are fewer flexible needle puncture straight paths, and more are a single arc path or a combination of multi-segment compound paths. The puncture path model Path of the flexible needle can be described as:

In the formula, Path is composed of n segments, p _i represents the starting point of the i-th path, θ _i represents the central angle of the i-th arc, r _i represents the radius of the i-th arc, if the path is a straight line, then r _i The value is 0, ω _i represents the arc tangent vector at the i-th path segment, and _Li represents the path function of the i-th segment.

To establish the path optimization objective function, it is necessary to follow the standard of the optimal path. It not only needs to enable the puncture needle to reach the target position accurately and safely, but also needs to minimize the patient's harm and pain. Thus, the path optimization objective function Y is determined as follows:

Y＝μ ₁ Y _n +μ ₂ Y _d +μ ₃ Y _l

In the formula, Y _n represents the limiting factor as a function of the error between the final puncture position and the target position, Y _d represents the limiting factor as a function of the distance between the trajectory and the obstacle, Y _l represents the limiting factor as a function of the path length, μ ₁ , μ ₂ , and μ ₃ are the weighting coefficients of Y _n , Y _d , and Y _l respectively, where μ ₁ > μ ₂ > μ ₃ shows the importance of the three factors. The relationships between the various expressions in the formula are as follows:

In the formula, l represents the allowed working length of the flexible needle; n represents the number of segments of the planned path; w represents the deviation between the final puncture position and the target position; w _max is the maximum allowed deviation between the final puncture position and the target position; l _i is the length of the path segment in the planned path; d is the minimum distance from obstacles in the planned path; d _min is the minimum allowed safe distance from obstacles.

Step (3) Population initialization operation, determine the control parameters of the differential evolution algorithm, randomly generate an initial population, and evaluate the initial population, that is, calculate the fitness value corresponding to each individual in the initial population. The initial population uses the following formula:

In the _{formula ,} and/> represents the limit of the parameter variable, rand(0,1) represents a random number in the interval [0,1], and NP represents the population size. It mainly reflects the amount of population information in the algorithm. The larger the NP value, the richer the population information is, but the amount of calculation will become larger, which is not conducive to solution. On the contrary, the smaller the NP is, the diversity of the population will be limited, which is not conducive to the algorithm finding the global optimal solution, and may even lead to search stagnation.

Step (4) Select individuals from the initial population according to the fitness value to perform mutation operations to obtain mutated individuals.

Specifically, two different individuals are selected from a randomly selected population, and their vector differences are scaled and then combined with the individual to be mutated to obtain the mutated individual, as shown in the following formula:

v _i (g)＝x _r1 (g)+F·(x _r2 (g)-x _r3 (g))

i≠r1≠r2≠r3

In the formula, x _r1 (g), x _r2 (g), x _r3 (g) represent three randomly selected individuals that are different from each other from the current group, and should not be the same as the target individual x _i , v _i (g ) represents the mutated individual corresponding to the target individual x _i (g), and F represents the scaling factor. F mainly affects the global optimization ability of the algorithm. The smaller F is, the better the algorithm's local search ability is. The larger F is, the better the algorithm can jump out of the local minimum point, but the convergence speed will be slower.

Step (5) Cross-operate the mutant individual with the target individual to obtain an experimental individual. Crossover operation: perform a crossover operation on the mutant individual and the target individual to obtain an experimental individual. The target individual is the product of the target number and the initial population.

Specifically, for each component, the offspring mutation vector is selected with a certain probability to generate experimental individuals.

In the formula, rand(0,1) represents a random number in the interval [0,1], j _rand represents an integer randomly selected from 1 to D, and CR represents the crossover operator. CR mainly reflects the degree of the amount of information exchanged between the test vector, the target vector, and the mutation vector during the crossover process. The larger the value of CR, the greater the degree of information exchange. On the contrary, if the value of CR is too small, the diversity of the population will decrease rapidly, which is not conducive to global optimization.

In practical applications, the mutation operation is followed by a crossover operation to increase the probability density.

Step (6) During the evolution process, perform boundary condition processing operations to obtain a temporary population. During the evolution process, it is judged whether each component in the mutant individual meets the boundary conditions. If the boundary conditions are not met, the mutant individual is regenerated using the generation method in step (3).

Step (7) Use a greedy strategy for the individuals in the temporary population and the target individual to select the optimal individual to form the target population.

The greedy strategy formula is as follows:

Specifically, using the above formula, the greedy strategy selects the better one from the target individual and the experimental individual as the next generation based on the value of the fitness function.

In practical applications, the purpose of performing selection operations is to decide which individuals to retain to the next generation or how many individuals to copy to the next generation. A greedy strategy selection scheme is applied in the differential evolution algorithm to determine the new individual u _i (g) and the target individual x _i (g). If u _i (g) is better than x _i (g), then the new individual will be replaced with the corresponding target vector in the next generation, otherwise the target individual remains in the target population. Therefore, the advantage of this approach is that the target population either gets better or remains in a healthy state, but never deteriorates.

Step (8) determines whether the termination condition is met or the maximum number of evolutions is reached: if so, the evolution is terminated and the best individual at this time is output as the solution; otherwise, steps (4) to (7) are repeated until the conditions are met.

The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and descriptions is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will also have various modifications. These changes and improvements all fall within the scope of the claimed invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims (9)

1. A flexible puncture needle path planning method based on differential evolution algorithm, which is characterized by: including the following steps: Step (1) Obtain the puncture environment image through the image collector, and determine the location of the target and obstacles from the image; Step (2) Establish the flexible needle puncture model and path optimization objective function; Step (3) Determine the control parameters of the differential evolution algorithm, randomly generate an initial population, and evaluate the initial population, that is, calculate the fitness value corresponding to each individual in the initial population; Step (4) Select individuals from the initial population according to the fitness value to perform mutation operations to obtain mutated individuals; Step (5) Perform a crossover operation on the mutant individual and the target individual to obtain an experimental individual; Step (6) During the evolution process, perform boundary condition processing operations to obtain a temporary population; Step (7) Use a greedy strategy for the individuals in the temporary population and the target individual to select the optimal individual to form the target population; Step (8) determines whether the termination condition is met or the maximum number of evolutions is reached: if so, the evolution is terminated and the best individual at this time is output as the solution; otherwise, steps (4) to (7) are repeated until the conditions are met. 2. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: establishing the flexible needle puncture model _Path in step (2) is as follows: In the formula, Path is composed of n segments, p _i represents the starting point of the i-th path, θ _i represents the central angle of the i-th arc, r _i represents the radius of the i-th arc, if the path is a straight line, then r _i The value is 0, ω _i represents the arc tangent vector at the i-th path segment, and _Li represents the path function of the i-th segment. 3. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: the path optimization objective function Y in step (2) is as follows: Y＝μ ₁ Y _n +μ ₂ Y _d +μ ₃ Y _l In the formula, Y _n represents the limiting factor as a function of the error between the final puncture position and the target position, Y _d represents the limiting factor as a function of the distance between the trajectory and the obstacle, Y _l represents the limiting factor as a function of the path length, μ ₁ , μ ₂ , and μ ₃ are the weighting coefficients of Y _n , Y _d , and Y _l respectively, where μ ₁ > μ ₂ > μ ₃ shows the importance of the three factors. 4. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 3, characterized in that: various relationships are as follows: In the formula, l represents the allowed working length of the flexible needle; n represents the number of segments of the planned path; w represents the deviation between the final puncture position and the target position; w _max is the maximum allowed deviation between the final puncture position and the target position; l _i is the length of the path segment in the planned path; d is the minimum distance from obstacles in the planned path; d _min is the minimum allowed safe distance from obstacles. 5. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: in step (3), the control parameters of the differential evolution algorithm are determined, the initial population is randomly generated, and the following formula is adopted: In the _{formula ,} and/> represents the limit of the parameter variable, rand(0,1) represents a random number in the interval [0,1], and NP represents the population size. 6. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: the mutation operation formula in step (4) is as follows: v _i (g)＝x _r1 (g)+F·(x _r2 (g)-x _r3 (g)) i≠r1≠r2≠r3 In the formula, x _r1 (g), x _r2 (g), x _r3 (g) represent three randomly selected individuals that are different from each other from the current group, and should not be the same as the target individual x _i , v _i (g ) represents the mutated individual corresponding to the target individual x _i (g), and F represents the scaling factor. 7. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: the crossover operation formula in step (5) is as follows: In the formula, rand(0,1) represents a random number in the interval [0,1], j _rand represents an integer randomly selected from 1 to D, and CR represents the crossover operator. 8. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: in step (6) during the evolution process, it is judged whether each component in the mutated individual satisfies the boundary condition. If not, If the boundary conditions are met, the mutant individuals will be regenerated using the generation method in step (3). 9. A flexible puncture needle path planning method based on differential evolution algorithm according to claim 1, characterized in that: the greedy strategy formula in step (7) is as follows: