Flexible needle dynamic obstacle avoidance path planning method based on improved BINN and optimized MOPSO
By combining the improved MOPSO and BINN algorithms with global and local path planning, the problems of puncture accuracy and safety of flexible needles in dynamic environments are solved, and efficient and accurate flexible needle puncture path planning is achieved.
Patent Information
- Application Number
- CN202511245212.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-02
- Publication Date
- 2025-12-12
AI Technical Summary
Most existing flexible needle puncture path planning algorithms are applicable to static environments and fail to effectively consider the influence of the flexible needle's own rotation angle on path planning. They are difficult to adapt to complex anatomical structures and dynamic changes, resulting in insufficient puncture accuracy and safety.
An improved multi-objective particle swarm optimization (MOPSO) and bio-inspired neural network (BINN) algorithm is used, combined with preoperative global path planning and intraoperative local path planning, to optimize the puncture path of the flexible needle and avoid fixed and moving obstacles through a dynamic obstacle avoidance path planning method.
It improves the adaptability and flexibility of flexible needle puncture, significantly enhances puncture accuracy and safety, and enables efficient and precise percutaneous puncture in complex and dynamic environments.
Smart Images

Figure CN121101747A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of flexible needle puncture path planning and dynamic obstacle avoidance technology, specifically to a flexible needle dynamic obstacle avoidance path planning method based on improved BINN and optimized MOPSO. Background Technology
[0002] Minimally invasive diagnostic and treatment techniques are widely used in clinical medicine. Compared with traditional surgery, their advantages include reduced tissue damage, reduced pain, and reduced risk of infection. Moreover, the technique is easy to operate and has a fast recovery time, making it an important treatment option. Percutaneous puncture surgery, guided by images, can perform tissue sampling, radiotherapy, drug infusion, and targeted ablation. However, existing imaging technologies have technical bottlenecks when dealing with complex dynamic environments, making it difficult to meet clinical safety requirements.
[0003] Traditional rigid puncture needles, limited by their material properties, can only achieve a near-straight puncture trajectory. In areas with complex anatomical structures and rich in vital tissues such as blood vessels and nerves, they struggle to safely reach the target without damaging surrounding tissues, greatly limiting their application. To address this deficiency, a new type of puncture tool, the flexible needle, has been developed. Due to its flexibility and asymmetrical tip angle, it can bend and deform when interacting with soft tissue. Due to the interaction force, the soft tissue exerts a non-uniform lateral force on the needle tip, while the flexible needle encounters cutting resistance perpendicular to the tip's bevel. The combined effect of these two forces causes the flexible needle to bend and deform, allowing it to advance along an arc path of a certain radius. By alternating between "insertion" and "rotation" modes, high-precision puncture is achieved, enabling obstacle avoidance, path adjustment, and real-time trajectory adjustment to accurately puncture the predetermined target location.
[0004] However, current procedures for puncturing with flexible needles face numerous challenges, including but not limited to: complex anatomical structures and unpredictable tissue deformation. Even minor errors during the procedure can lead to serious consequences. Therefore, to achieve precise puncture, advanced path planning techniques have been proposed to ensure that the flexible needle accurately reaches the target location in a dynamic environment while avoiding collisions with obstacles. Existing technologies typically employ algorithms such as A* (A-Star) and RRT (Rapidly-exploring Random Tree). However, these algorithms do not consider the impact of the flexible needle's own rotation angle on path planning during puncture and are mostly applicable to static environments, making it difficult to adapt to dynamic changes that may occur in actual operations. This results in limitations in real-time dynamic obstacle avoidance. Summary of the Invention
[0005] To address the technical problem that most existing path planning algorithms for flexible needles are applicable to static environments and do not consider the influence of the flexible needle's own rotation angle on path planning, thus failing to adapt to environmental changes, the present invention aims to provide a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO. The specific technical solution adopted is as follows:
[0006] Obtain the puncture site, initialize the puncture environment, and update the environmental parameters in the puncture environment;
[0007] Based on environmental parameters, determine whether there are fixed obstacles in the current puncture path of the flexible needle. If so, use optimized MOPSO to avoid fixed obstacles and determine the global optimal path.
[0008] Determine if there are active obstacles on the globally optimal path. If so, improve the BINN to avoid the active obstacles and output the final puncture path.
[0009] Preferably, determining whether there are fixed obstacles along the current puncture path of the flexible needle based on environmental parameters includes:
[0010] If so, optimize MOPSO to avoid fixed obstacles and determine the globally optimal path, specifically as follows:
[0011] Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system and to establish a flexible needle model.
[0012] Construct a path planning objective function, dynamically weight and optimize MOPSO, avoid fixed obstacles, and output the globally optimal path;
[0013] If not, directly calculate the path planning objective function and output the globally optimal path.
[0014] Preferably, the environmental parameters include the curvature of the flexible needle puncture, the needle tail advance speed, and the rotational angular velocity.
[0015] Preferably, the flexible needle is mapped to the world coordinate system based on environmental parameters to determine the position and orientation of the needle tip in the world coordinate system, and a flexible needle model is established, including:
[0016] Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system, denoted as . ,in, , , They represent circumference respectively. , , The rotation angle of the shaft;
[0017] The velocity of the flexible needle in the world coordinate system is obtained. The flexible needle moves in an arc along its curvature. A needle body coordinate system is constructed based on the flexible needle, and the velocity in this coordinate system is determined. The conversion relationship between the velocity in the world coordinate system and the velocity in the needle body coordinate system is obtained, and the corresponding calculation formula is as follows:
[0018]
[0019]
[0020]
[0021]
[0022]
[0023] in, This indicates the velocity of the flexible needle in the world coordinate system; The derivative operation represents the position and orientation of the tip of a flexible needle in the world coordinate system; This indicates the velocity of the flexible needle in the needle body coordinate system; Indicates the advance speed of the tail needle of the flexible needle; Indicates the rotational angular velocity of the flexible needle; , These represent the propulsion and rotation speeds of the flexible needle, respectively. Indicates curvature; This represents the angular velocity that converts the velocity in the world coordinate system to the velocity in the needle coordinate system. This represents the rotation matrix between the needle coordinate system and the world coordinate system. The matrix coefficients of the Euler kinematic equations representing the fixed-point motion of the flexible needle;
[0024] A flexible needle model is established based on the transformation relationship.
[0025] Preferably, a flexible needle model is established based on the conversion relationship, and the corresponding calculation formula is as follows:
[0026]
[0027] in, This represents the rate of change of the pose of the flexible needle over time. express The differentiation operation; , , They represent circumference respectively. , , The rotation angle of the shaft; Indicates curvature; Indicates the advance speed of the tail needle of the flexible needle; This indicates the rotational angular velocity of the flexible needle.
[0028] Preferably, a path planning objective function is constructed, MOPSO is dynamically and weighted to avoid fixed obstacles, and the globally optimal path is output, including:
[0029] Based on the current puncture path, the path deviation, path length, and path risk are determined. A path planning objective function is constructed by combining this with the total rotation angle of the flexible needle, and then minimized. The corresponding calculation formula is as follows:
[0030]
[0031] in, Represent the objective function for path planning; This indicates minimization. Indicates path deviation; Indicates the path length; Indicates the hazard level of the route; Indicates the total rotation angle of the flexible needle; Indicates the decision quantity;
[0032] Based on the weighted processing of the path planning objective function, MOPSO is optimized to avoid fixed obstacles and output the globally optimal path.
[0033] Preferably, determining whether there are active obstacles on the globally optimal path includes:
[0034] If so, improve the BINN to avoid moving obstacles and output the final puncture path, specifically as follows:
[0035] The puncture is performed based on the global optimal path. If there is a moving obstacle in front, the distance between the current position of the flexible needle and the boundary of the moving obstacle is calculated to determine whether they intersect. If they do, all path points and moving obstacles in the global optimal path are traversed to identify all intersecting path points.
[0036] Based on the intersection of path points, check whether they are continuous with the previous adjacent intersection path point, filter out continuous intersection path segments, and construct an intersection point sequence;
[0037] Using BINN and improving local path planning for the intersection point sequence, avoiding moving obstacles, merging local paths and the global optimal path, and outputting the final puncture path;
[0038] If not, output the globally optimal path as the final puncture path.
[0039] Preferably, the distance between the current position of the flexible needle and the boundary of the moving obstacle is calculated to determine whether they intersect, specifically as follows:
[0040] Set a safe distance, denoted as... The distance between the pathpoint and the boundary of the active obstacle is calculated and denoted as... ,in, Indicates the first 1 path point; Indicates the first One obstacle to movement; when When an obstacle is encountered, the path point is determined to intersect with the obstacle.
[0041] Preferably, BINN is used and improved local path planning is performed on the intersection point sequence to avoid moving obstacles. The local path and the global optimal path are merged to output the final puncture path, including:
[0042] Based on the environment at the current position of the flexible needle, a local model is performed using BINN. Based on the connection between each neuron and its neighboring neurons in the BINN, a transmission network is constructed. The flexible needle is placed in the transmission network, and the rate of change of neuron state is calculated. The neighboring neurons with the highest neuron activity value are selected to determine the puncture direction. Moving obstacles are avoided, and a local path is established until there are no moving obstacles in front of the current position of the flexible needle. The global optimal path is then returned. The local path and the global optimal path are merged, and the final puncture path is output.
[0043] Preferably, the process includes calculating the rate of change of neuronal state, selecting neighboring neurons with the highest neuronal activity values to determine the puncture direction, avoiding obstacles, and establishing a local path, including:
[0044] Define the neuron currently occupied by the flexible needle as the first neuron, and the neighboring neurons occupied by the moving obstacle as the second neuron. Calculate the rate of change of neuron state using the following formula:
[0045]
[0046]
[0047]
[0048] in, Indicates the first Neurons over time The rate of change of state; Indicates the first The first neuron; This represents the natural decay rate of the neuron's activity value over time. , These represent the upper and lower limits of neuronal activity values, respectively. Indicates the first The excitatory stimulation of each neuron; Indicates the first Inhibitory stimulation of individual neurons; This represents the state information of the first neuron; This indicates the state information of the second neuron; Indicates the first A neuron from other neurons The sum of excitatory stimuli received;
[0049] Similarly, determine the activity values of all neurons in the transmission network and sort them in descending order. Based on the first neuron, select the neighboring neurons with the highest activity values as the puncture direction, avoid activity obstacles, and establish a local path.
[0050] The present invention has the following beneficial effects:
[0051] This paper proposes a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO, combining the Multiple Objective Particle Swarm Optimization (MOPSO) algorithm with an improved Bioinspired Neural Network (BINN) algorithm. This method enables efficient, accurate percutaneous punctures with dynamic obstacle avoidance capabilities. Considering multiple key parameters such as path length, risk level, number of needle rotations, and needle rotation angle, the BINN algorithm is used to detect and avoid moving obstacles that may occur during the puncture process in real time. By combining MOPSO for preoperative global path planning and BINN for intraoperative local planning, the method fully leverages the comprehensiveness of preoperative planning and the real-time nature of intraoperative planning, effectively balancing the various needs of path planning. This not only improves the adaptability and flexibility of flexible needle punctures but also provides reliable path planning support for complex puncture procedures, significantly enhancing puncture accuracy and safety, and possessing significant clinical application value. Attached Figure Description
[0052] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0053] Figure 1 The following is a flowchart illustrating an implementation method for a dynamic obstacle avoidance path planning method for flexible needles based on an improved BINN and optimized MOPSO, as provided in one embodiment of the present invention.
[0054] Figure 2A flexible needle model diagram provided in an embodiment of the present invention, which is based on an improved BINN and an optimized MOPSO flexible needle dynamic obstacle avoidance path planning method;
[0055] Figure 3 A path planning illustration of a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO, provided as an embodiment of the present invention. Figure 1 ;
[0056] Figure 4 A path planning illustration of a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO, provided as an embodiment of the present invention. Figure 2 ;
[0057] Figure 5 A path planning illustration of a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO, provided as an embodiment of the present invention. Figure 3 . Detailed Implementation
[0058] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0060] The following description, in conjunction with the accompanying drawings, details a specific scheme for a dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO.
[0061] Please see Figure 1 The diagram illustrates a flowchart of a dynamic obstacle avoidance path planning method for flexible needles based on an improved BINN and optimized MOPSO according to an embodiment of the present invention. The method includes:
[0062] Step S1: Obtain the puncture site, initialize the puncture environment, and update the environmental parameters in the puncture environment;
[0063] Step S2: Determine whether there are fixed obstacles in the current puncture path of the flexible needle based on environmental parameters. If so, use optimized MOPSO to avoid fixed obstacles and determine the global optimal path.
[0064] Step S3: Determine if there are active obstacles in the global optimal path. If so, improve the BINN to avoid the active obstacles and output the final puncture path.
[0065] To better illustrate this point, traditional rigid needles, limited by their material properties, can only achieve straight-line punctures. While existing flexible needles can achieve curved punctures, current puncture algorithms fail to consider the impact of needle rotation angle on path planning and tissue damage. This leads to deviations from the intended trajectory in complex tissue structures, increasing surgical risks. Furthermore, their path planning is typically designed for static environments and cannot meet the dynamic and unexpected conditions encountered during actual intraoperative punctures, such as tissue fragments, air bubbles, organ movement, or changes in tumor location, thus reducing clinical safety. Therefore, a dynamic obstacle avoidance path planning method based on improved BINN and optimized MOPSO flexible needles is proposed. This method combines preoperative global path planning and intraoperative local planning to balance path planning requirements, improve puncture accuracy, and enhance surgical safety and success rates. Among these methods, the MOPSO multi-objective particle swarm optimization algorithm solves complex optimization problems by simulating the collective behavior of organisms in nature. BINN, derived from neural membrane circuit models and action potential transmission formulas, combines the characteristics of biological neural networks with the efficient computational power of electronic circuits, enabling faster and more accurate information processing and transmission.
[0066] It can be explained that in step S1, the puncture site is obtained, the puncture environment is initialized, and the environmental parameters in the puncture environment are updated. That is, at this time, preparations are made for the puncture operation at the puncture site. The condition of the skin and tissue at the puncture site is carefully checked to ensure that there are no signs of infection. Then, the puncture environment is initialized, including disinfecting the area, preparing sterile instruments, and adjusting the lighting to provide the best vision. Then, the environmental parameters in the puncture environment are updated. Preferably, through a dynamic feedback mechanism, the speed and position are optimized according to the real-time data of particles or neurons to truly reflect the puncture environment and ensure the reliability of subsequent path planning.
[0067] Understandably, when planning the global path before surgery based on the puncture site, it is necessary to determine whether there are fixed obstacles such as bones, blood vessels or other important tissue structures along the puncture path. This is to ensure that the obtained global optimal path avoids fixed obstacles and does not collide with any fixed obstacles, thereby minimizing damage to surrounding healthy tissues, improving the success rate of the surgery, and thus improving the postoperative recovery effect.
[0068] Further, in step S2, determining whether there are fixed obstacles along the current puncture path of the flexible needle based on environmental parameters includes:
[0069] Step S21: If so, use optimized MOPSO to avoid fixed obstacles and determine the globally optimal path, specifically:
[0070] Step S211: Based on environmental parameters, map the flexible needle to the world coordinate system, determine the position and orientation of the needle tip in the world coordinate system, and establish a flexible needle model.
[0071] It can be explained that the flexible needle is subjected to asymmetrical forces by the punctured tissue during the puncture process, which causes the needle tip to deflect. Therefore, in order to control the rotation of the needle tail to deflect the needle tip and travel along the puncture path, a flexible needle model is established.
[0072] Flexible needle models are divided into dynamic models and kinematic models. Dynamic models usually rely on static analysis, making it difficult to effectively simulate the multiple rotational movements that may occur during the insertion of the flexible needle. Furthermore, dynamic models are relatively complex and mostly remain at the stage of theoretical discussion and analysis. Kinematic models, on the other hand, do not require mechanical analysis of the interaction forces and friction between the flexible needle tip and soft tissue. Instead, they are directly constructed based on experimental data of the flexible needle's puncture trajectory within the soft tissue. Patterns are extracted directly from the trajectory data, such as the motion of a fixed-curvature circular arc and the splicing of multiple circular arcs. There is no need to establish a correlation between force and needle tip deflection or trajectory changes, simplifying the complexity of the interaction between the flexible needle and soft tissue and making the kinematic model structure more concise. Optionally, In this embodiment, the flexible needle model is based on the kinematic model proposed by Patil et al., which states that the flexible needle moves in an arc along a fixed curvature within the tissue. By rotating the tail of the flexible needle, the direction of the needle tip can be adjusted accordingly, thereby changing the needle path. Therefore, the following assumptions are made regarding the puncture: First, when the tail of the flexible needle is rotated, the needle tip can immediately follow the rotation direction of the needle tail and maintain the same orientation as the needle tail; Second, the puncture site, i.e., the soft tissue to be punctured, has a uniform density and does not undergo any form of deformation during the puncture; Third, the flexible needle body moves completely according to the movement trajectory of the needle tip and does not exhibit radial drift after entering the soft tissue. Then, based on the above assumptions, the puncture site is planned.
[0073] Furthermore, the environmental parameters include the curvature of the flexible needle puncture, the needle tail advancement speed, and the rotational angular velocity. It should be noted that these also include the velocity and location of particles or neurons in the puncture site. The curvature determines the degree of bending and path of the needle tip within the tissue; the needle tail advancement speed refers to the speed at which the flexible needle tail advances forward, affecting the depth and speed at which the needle enters the tissue; and the rotational angular velocity represents the speed at which the needle rotates around its own axis during its advancement. Analyzing these environmental parameters helps adjust the direction of the flexible needle and reduce damage to surrounding tissues.
[0074] Please see Figure 2It shows a flexible needle model diagram of a flexible needle dynamic obstacle avoidance path planning method based on improved BINN and optimized MOPSO provided in the first embodiment of the present invention; wherein, the tip of the flexible needle follows a fixed curvature. The needle moves in a circular motion, with the dotted line representing the radius of the arc. Adjusting the advance speed and rotational angular velocity at the needle's tail will cause the needle tip to move from its current position. Move to the next position .
[0075] Further, step S211 includes:
[0076] Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system, denoted as . ,in, , , They represent circumference respectively. , , The rotation angle of the shaft;
[0077] The velocity of the flexible needle in the world coordinate system is obtained. The flexible needle moves in an arc along its curvature. A needle body coordinate system is constructed based on the flexible needle, and the velocity in this coordinate system is determined. The conversion relationship between the velocity in the world coordinate system and the velocity in the needle body coordinate system is obtained, and the corresponding calculation formula is as follows:
[0078]
[0079]
[0080]
[0081]
[0082]
[0083] in, This indicates the velocity of the flexible needle in the world coordinate system; The derivative operation represents the position and orientation of the tip of a flexible needle in the world coordinate system; This indicates the velocity of the flexible needle in the needle body coordinate system; Indicates the advance speed of the tail needle of the flexible needle; Indicates the rotational angular velocity of the flexible needle; , These represent the propulsion and rotation speeds of the flexible needle, respectively. Indicates curvature; This represents the angular velocity that converts the velocity in the world coordinate system to the velocity in the needle coordinate system. This represents the rotation matrix between the needle coordinate system and the world coordinate system. The matrix coefficients of the Euler kinematic equations representing the fixed-point motion of the flexible needle;
[0084] A flexible needle model is established based on the transformation relationship.
[0085] It is explained that when performing punctures within soft tissue, one will feel the resistance and elasticity of the soft tissue, and the tip of the flexible needle follows the curvature. The circular trajectory motion of the flexible needle is determined by the position of the needle tip and the orientation of the inclined plane, thereby determining the position and attitude of the flexible needle tip in the world coordinate system, establishing the mapping relationship between the world coordinate system and the flexible needle body coordinate system, and establishing the kinematic model of the flexible needle.
[0086] Furthermore, a flexible needle model is established based on the transformation relationship, and the corresponding calculation formula is as follows:
[0087]
[0088] in, This represents the rate of change of the pose of the flexible needle over time. express The differentiation operation; , , They represent circumference respectively. , , The rotation angle of the shaft; Indicates curvature; Indicates the advance speed of the tail needle of the flexible needle; This indicates the rotational angular velocity of the flexible needle.
[0089] It is explained that the flexible needle model is a non-holonomic system with input and state coupling, making direct puncture path planning and control difficult. Furthermore, related studies indicate that the flexible needle puncture trajectory is independent of speed and angular velocity, allowing for the separation of time factors. By using needle insertion distance and rotation angle as control variables, the flexible needle model is simplified, and based on this, research on optimizing flexible needle path planning is conducted to ensure the stability of preoperative global path planning.
[0090] Step S212: Construct the path planning objective function, dynamically weight and optimize MOPSO, avoid fixed obstacles, and output the globally optimal path.
[0091] Understandably, MOPSO, or Multi-Objective Particle Swarm Optimization, can solve optimization problems between multiple conflicting objective functions. In existing technologies, only the factors of the puncture path itself are usually considered. However, in actual surgical operations, the rotation angle of the flexible needle is also closely related to the patient's condition. That is, the larger the rotation angle of the flexible needle, the more severe the tissue damage to the patient. Therefore, in order to ensure improved safety, the magnitude of the rotation angle of the flexible needle each time is considered to ensure the comprehensiveness of the overall planning.
[0092] Further, step S212 includes:
[0093] Based on the current puncture path, the path deviation, path length, and path risk are determined. A path planning objective function is constructed by combining this with the total rotation angle of the flexible needle, and then minimized. The corresponding calculation formula is as follows:
[0094]
[0095] in, Represent the objective function for path planning; This indicates minimization. Indicates path deviation; Indicates the path length; Indicates the hazard level of the route; Indicates the total rotation angle of the flexible needle; Indicates the decision quantity;
[0096] Based on the weighted processing of the path planning objective function, MOPSO is optimized to avoid fixed obstacles and output the globally optimal path.
[0097] The procedure involves defining the puncture site as the target point and performing a puncture through the skin. Here, path deviation represents the distance between the last path point and the target point in the planned path; path length represents the sum of distances from all path points; and path hazard represents the sum of distances from the path to fixed obstacles. A path planning objective function is then established. In practical applications, all four objectives need to be optimized simultaneously. However, there is a contradiction between path deviation and path length: pursuing the shortest path leads to the maximum deviation, while pursuing a zero-deviation path results in a longer path length. To resolve this contradiction, the path planning objective function is weighted, dynamically adjusting the weight coefficients for path deviation and path length at different stages. This optimizes MOPSO and improves the optimization of path deviation and path length. The corresponding calculation formula is as follows:
[0098]
[0099] in, , These represent the corresponding weight coefficients for path deviation and path length, respectively; thus, the goal of avoiding fixed obstacles is achieved, and the path is determined as the globally optimal path and output.
[0100] As an optional implementation, if there is only one fixed obstacle in the current puncture path, and the distance from the flexible needle to the fixed obstacle is less than the safe distance... If the path hazard is infinite, then the path hazard is infinite; conversely, if it is greater than the safe distance, then the path hazard is... ,in, Indicates the safety factor. This represents the shortest straight-line distance from the flexible needle to the fixed obstacle; and when there are multiple fixed obstacles, the path hazard level is... The sum of these factors; it can be explained that the safety distance refers to the minimum safe interval set in actual operation to ensure that the flexible needle does not collide with obstacles; the safety factor is a dimensionless value used to adjust the calculation results to take into account uncertainties and risk factors.
[0101] Step S22: If not, directly calculate the path planning objective function and output the globally optimal path; that is, when planning the preoperative puncture path for the puncture site, there are no fixed obstacles in the determined puncture path. In this case, the globally optimal path is output directly through the path planning objective function based on the path planning decision objective, thereby reducing surgical risks.
[0102] Further, in step S3, determining whether there are active obstacles on the globally optimal path includes:
[0103] Understandably, the globally optimal path is based on the puncture path determined before the puncture surgery. When the flexible needle is punctured to a certain position in the site according to the previously planned path, and there are moving obstacles such as tissue fragments or air bubbles in front, and there is a path point that intersects with the moving obstacle, BINN is used to plan the local path to ensure safety and accuracy during the operation.
[0104] Step S31: If yes, improve the BINN to avoid moving obstacles and output the final puncture path, specifically:
[0105] Step S311: Perform puncture based on the global optimal path. If there is a moving obstacle in front, calculate the distance between the current position of the flexible needle and the boundary of the moving obstacle to determine whether they intersect. If they do, traverse all path points and moving obstacles in the global optimal path to identify all intersecting path points.
[0106] Further, in step S311, the distance between the current position of the flexible needle and the boundary of the moving obstacle is calculated to determine whether they intersect, specifically as follows:
[0107] Set a safe distance, denoted as... The distance between the pathpoint and the boundary of the active obstacle is calculated and denoted as... ,in, Indicates the first 1 path point; Indicates the first One obstacle to movement; when When an obstacle is encountered, the path point is determined to intersect with the obstacle.
[0108] It is explained that the safety distance is used to deal with unexpected situations during the puncture process. If the distance between the path point and the moving obstacle is less than the safety distance, it means that the path point and the moving obstacle intersect. In this case, the puncture path needs to be adjusted during local path planning to avoid collision.
[0109] Step S312: Check whether the intersecting path points are continuous with the adjacent previous intersecting path point, filter out continuous intersecting path segments, and construct an intersection point sequence; that is, if the distance between the adjacent previous intersecting path point and the moving obstacle is less than or equal to the safe distance, it means that there is a continuous dangerous intersecting path segment. Filter it out and construct an intersection point sequence to facilitate subsequent local path planning to avoid the moving obstacle.
[0110] Step S313: Use BINN and improve local path planning for the intersection point sequence to avoid moving obstacles, merge local paths and the global optimal path, and output the final puncture path.
[0111] Further, step S313 includes:
[0112] Based on the environment at the current position of the flexible needle, a local model is performed using BINN. Based on the connection between each neuron and its neighboring neurons in the BINN, a transmission network is constructed. The flexible needle is placed in the transmission network, and the rate of change of neuron state is calculated. The neighboring neurons with the highest neuron activity value are selected to determine the puncture direction. Moving obstacles are avoided, and a local path is established until there are no moving obstacles in front of the current position of the flexible needle. The global optimal path is then returned. The local path and the global optimal path are merged, and the final puncture path is output.
[0113] It can be explained that when performing local path planning based on any segment of continuously intersecting path points in the intersection point sequence, that is, performing local modeling based on the current puncture environment, a system composed of neurons is established in the three-dimensional local space. The mesh diagram shows that each neuron represents a potential puncture direction. In the local space, based on the current position of the flexible needle, there are 17 feasible directions, which are composed of substitute neurons and reference neurons. The substitute neurons are neurons that can be replaced in the direction of the flexible needle puncture, while the reference neurons are neurons that can move toward the flexible needle puncture path. The state information of each neuron is then determined.
[0114] Specifically, if the reference neuron behind the surrogate neuron is occupied by an obstacle, the activity information of the surrogate neuron will be reduced, and the flexible needle will avoid moving in that direction. The corresponding state information is defined as follows:
[0115]
[0116]
[0117] in, This indicates the state information of the surrogate neuron; This indicates the state information of the reference neuron; A scalar representing the value of neuronal activity, used to quantify the intensity of neuronal activity and to distinguish the degree of influence of different locations, such as target location and obstacle location, on neuronal activity.
[0118] To clarify, the target location refers to the site to be punctured; the movable location refers to the direction in which the flexible needle can currently travel during puncture; the obstacle location is the location of the moving obstacle; and the flexible needle location is the current location of the neuron.
[0119] Furthermore, the rate of change of neuronal state is calculated, and neighboring neurons with the highest neuronal activity values are selected to determine the puncture direction, avoiding obstacles and establishing a local path, including:
[0120] It can be explained that the activity information of each neuron in the local modeling is updated in real time according to the puncture environment, and each neuron is connected to neighboring neurons to form an active transmission network. In this embodiment, the current position of the flexible needle is defined as the center of the local modeling to effectively detect the environment within the local area and improve the transmission efficiency of neuronal information.
[0121] Define the neuron currently occupied by the flexible needle as the first neuron, and the neighboring neurons occupied by the moving obstacle as the second neuron. Calculate the rate of change of neuron state using the following formula:
[0122]
[0123]
[0124]
[0125] in, Indicates the first Neurons over time The rate of change of state; Indicates the first The first neuron; This represents the natural decay rate of the neuron's activity value over time. , These represent the upper and lower limits of neuronal activity values, respectively. Indicates the first The excitatory stimulation of each neuron; Indicates the first Inhibitory stimulation of individual neurons; This represents the state information of the first neuron; This indicates the state information of the second neuron; Indicates the first A neuron from other neurons The sum of excitatory stimuli received;
[0126] Similarly, determine the activity values of all neurons in the transmission network and sort them in descending order. Based on the first neuron, select the neighboring neurons with the highest activity values as the puncture direction, avoid activity obstacles, and establish a local path.
[0127] To clarify, the first neuron is the substitute neuron; the second neuron is the reference neuron. By analyzing the neurons, the rate of change of each neuron's state is determined, and thus the activity value corresponding to each neuron is obtained. When a moving obstacle appears in front of the current position of the flexible needle, the neighboring neuron with the highest activity value is selected as the puncture direction to avoid the moving obstacle. If the moving obstacle keeps moving in front, BINN is used to repeatedly perform local planning until the moving obstacle is bypassed to avoid collisions between the two and unnecessary damage, thus obtaining a local path and regressing to the global optimal path. Finally, the local path and the global optimal path are integrated to output the final puncture path of the flexible needle, that is, BINN avoids the disadvantage of getting trapped in local optimization.
[0128] Step S32: If not, output the global optimal path as the final puncture path; that is, if there are no moving obstacles in the currently determined global optimal path of the flexible needle, then the final puncture path determined before the operation will be used directly as the final puncture path.
[0129] It is explained that combining preoperative global path planning with intraoperative local planning can make full use of the comprehensive preoperative planning and the real-time adjustment capabilities during the operation. That is, preoperative planning ensures the optimality and safety of the puncture path, while intraoperative local planning is used to deal with dynamic changes, ensure the smooth progress of the operation, and improve the efficiency and success rate of the operation.
[0130] Understandably, by combining multi-objective particle swarm optimization (MPS) algorithms with improved bio-inspired neural network (BINN) algorithms, efficient, accurate percutaneous punctures with dynamic obstacle avoidance capabilities can be achieved. Considering multiple key parameters such as path length, risk level, number of needle rotations, and needle rotation angle, the BINN algorithm is used to detect and avoid moving obstacles that may occur during the puncture process in real time. That is, by combining MOPSO for preoperative global path planning and BINN for intraoperative local planning, the advantages of comprehensive preoperative planning and real-time intraoperative planning are fully utilized, effectively balancing the various needs of path planning. This not only improves the adaptability and flexibility of flexible needle puncture but also provides reliable path planning support for complex puncture procedures, significantly improving puncture accuracy and safety, and has significant clinical application value.
[0131] To better illustrate and verify the effectiveness of the method in this application, two-dimensional and three-dimensional simulation experiments were conducted based on flexible needles.
[0132] Please combine Figure 3 and Figure 4 The experiment results of a two-dimensional simulation are shown. Specifically, in the two-dimensional plane, the initial point of the flexible needle, i.e., the needle insertion point, is assumed to be (0, 0), and the target point, i.e., the puncture site, is (1200, 1200). Three coordinates, (350, 100), (200, 600), and (800, 300), are set to simulate static obstacles during preoperative puncture path planning. After the optimized MOPSO algorithm is used to output the globally optimal path, two additional coordinates, (600, 900) and (500, 300), are added to simulate moving obstacles during the puncture process. The improved BINN algorithm is then used for local path planning on the globally optimal path. It can be observed that when moving obstacles appear on the globally optimal path, the BINN algorithm responds quickly and adjusts to avoid them. The algorithm makes reasonable bending and turning adjustments to avoid moving obstacles, without deviating from the general direction of the global plan, and successfully avoids the moving obstacles. This demonstrates the effectiveness of the improved BINN algorithm. The algorithm can achieve real-time obstacle avoidance correction in a two-dimensional dynamic environment without disrupting the globally optimal path planned by MOPSO.
[0133] Please see Figure 5The experiment results of the three-dimensional simulation are shown. Specifically, in three-dimensional space, the initial point of the flexible needle, i.e., the needle insertion point, is (0, 0, 0), and the target point, i.e. the puncture site, is (120, 120, 120). Similarly, three coordinates (20, 70, 40), (80, 35, 45), and (75, 95, 90) are set to simulate fixed obstacles. After the global optimal path is generated, two coordinates (44, -12, 55) and (60, 30, 30) are added to simulate moving obstacles to approximate or appear on the global optimal path. It can be seen that the BINN algorithm makes reasonable bending and deflection adjustments for moving obstacles, without deviating from the general direction of the global path planned by MOPSO, and successfully avoids the newly added moving obstacles. That is, the improved BINN algorithm can achieve real-time correction and obstacle avoidance in a three-dimensional dynamic environment without destroying the global path planned by MOPSO.
[0134] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0135] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO, characterized in that, The method includes: Obtain the puncture site, initialize the puncture environment, and update the environmental parameters in the puncture environment; Based on environmental parameters, determine whether there are fixed obstacles in the current puncture path of the flexible needle. If so, use optimized MOPSO to avoid fixed obstacles and determine the global optimal path. Determine if there are active obstacles on the globally optimal path. If so, improve the BINN to avoid the active obstacles and output the final puncture path.
2. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 1, characterized in that, Determining whether there are fixed obstacles along the current puncture path of the flexible needle based on environmental parameters, including: If so, optimize MOPSO to avoid fixed obstacles and determine the globally optimal path, specifically as follows: Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system and to establish a flexible needle model. Construct a path planning objective function, dynamically weight and optimize MOPSO, avoid fixed obstacles, and output the globally optimal path; If not, directly calculate the path planning objective function and output the globally optimal path.
3. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 2, characterized in that, The environmental parameters include the curvature of the flexible needle puncture, the needle tail advance speed, and the rotational angular velocity.
4. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 3, characterized in that, Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system, and a flexible needle model is established, including: Based on environmental parameters, the flexible needle is mapped to the world coordinate system to determine the position and orientation of the needle tip in the world coordinate system, denoted as . ,in, , , They represent circumference respectively. , , The rotation angle of the shaft; The velocity of the flexible needle in the world coordinate system is obtained. The flexible needle moves in an arc along its curvature. A needle body coordinate system is constructed based on the flexible needle, and the velocity in this coordinate system is determined. The conversion relationship between the velocity in the world coordinate system and the velocity in the needle body coordinate system is obtained, and the corresponding calculation formula is as follows: ; ; ; ; ; in, This indicates the velocity of the flexible needle in the world coordinate system; The derivative operation represents the position and orientation of the tip of a flexible needle in the world coordinate system; This indicates the velocity of the flexible needle in the needle body coordinate system; Indicates the advance speed of the tail needle of the flexible needle; Indicates the rotational angular velocity of the flexible needle; , These represent the propulsion and rotation speeds of the flexible needle, respectively. Indicates curvature; This represents the angular velocity that converts the velocity in the world coordinate system to the velocity in the needle coordinate system. This represents the rotation matrix between the needle coordinate system and the world coordinate system. The matrix coefficients of the Euler kinematic equations representing the fixed-point motion of the flexible needle; A flexible needle model is established based on the transformation relationship.
5. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 4, characterized in that, Based on the transformation relationship, a flexible needle model is established, and the corresponding calculation formula is: ; in, This represents the rate of change of the pose of the flexible needle over time. express The differentiation operation; , , They represent circumference respectively. , , The rotation angle of the shaft; Indicates curvature; Indicates the advance speed of the tail needle of the flexible needle; This indicates the rotational angular velocity of the flexible needle.
6. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 2, characterized in that, Construct a path planning objective function, dynamically weighted optimize MOPSO, avoid fixed obstacles, and output the globally optimal path, including: Based on the current puncture path, the path deviation, path length, and path risk are determined. A path planning objective function is constructed by combining this with the total rotation angle of the flexible needle, and then minimized. The corresponding calculation formula is as follows: ; in, Represent the objective function for path planning; This indicates minimization. Indicates path deviation; Indicates the path length; Indicates the hazard level of the route; Indicates the total rotation angle of the flexible needle; Indicates the decision quantity; Based on the weighted processing of the path planning objective function, MOPSO is optimized to avoid fixed obstacles and output the globally optimal path.
7. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 1, characterized in that, Determine if there are active obstacles on the globally optimal path, including: If so, improve the BINN to avoid moving obstacles and output the final puncture path, specifically as follows: The puncture is performed based on the global optimal path. If there is a moving obstacle in front, the distance between the current position of the flexible needle and the boundary of the moving obstacle is calculated to determine whether they intersect. If they do, all path points and moving obstacles in the global optimal path are traversed to identify all intersecting path points. Based on the intersection of path points, check whether they are continuous with the previous adjacent intersection path point, filter out continuous intersection path segments, and construct an intersection point sequence; Using BINN and improving local path planning for the intersection point sequence, avoiding moving obstacles, merging local paths and the global optimal path, and outputting the final puncture path; If not, output the globally optimal path as the final puncture path.
8. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 7, characterized in that, Calculate the distance between the current position of the flexible needle and the boundary of the moving obstacle to determine whether they intersect. Specifically: Set a safe distance, denoted as... The distance between the pathpoint and the boundary of the active obstacle is calculated and denoted as... ,in, Indicates the first 1 path point; Indicates the first One obstacle to movement; when When an obstacle is encountered, the path point is determined to intersect with the obstacle.
9. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 7, characterized in that, Using BINN and improving local path planning for intersection point sequences to avoid moving obstacles, merging local paths and the global optimal path, the final puncture path is output, including: Based on the environment at the current position of the flexible needle, a local model is performed using BINN. Based on the connection between each neuron and its neighboring neurons in the BINN, a transmission network is constructed. The flexible needle is placed in the transmission network, and the rate of change of neuron state is calculated. The neighboring neurons with the highest neuron activity value are selected to determine the puncture direction. Moving obstacles are avoided, and a local path is established until there are no moving obstacles in front of the current position of the flexible needle. The global optimal path is then returned. The local path and the global optimal path are merged, and the final puncture path is output.
10. The dynamic obstacle avoidance path planning method for flexible needles based on improved BINN and optimized MOPSO as described in claim 9, characterized in that, Calculate the rate of change of neuronal state, screen neighboring neurons with the highest neuronal activity values to determine the puncture direction, avoid obstacles, and establish a local pathway, including: Define the neuron currently occupied by the flexible needle as the first neuron, and the neighboring neurons occupied by the moving obstacle as the second neuron. Calculate the rate of change of neuron state using the following formula: ; ; ; in, Indicates the first Neurons over time The rate of change of state; Indicates the first The first neuron; This represents the natural decay rate of the neuron's activity value over time. , These represent the upper and lower limits of neuronal activity values, respectively. Indicates the first The excitatory stimulation of each neuron; Indicates the first Inhibitory stimulation of individual neurons; This represents the state information of the first neuron; This indicates the state information of the second neuron; Indicates the first A neuron from other neurons The sum of excitatory stimuli received; Similarly, determine the activity values of all neurons in the transmission network and sort them in descending order. Based on the first neuron, select the neighboring neurons with the highest activity values as the puncture direction, avoid activity obstacles, and establish a local path.
Citation Information
Cited By
Assembling path planning device and method for flexible sensing chip for textile auxiliary materials
CN121998220A
An assembly path planning device and method for flexible sensor chips used in textile accessories
CN121998220B
