A robot dynamic tracking and following positioning method and device for brain stimulation

Abstract

This invention relates to a method and apparatus for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation. The method includes establishing a spatial mapping relationship between multiple coordinate systems, transforming the coordinates of the stimulation target point and the initial cortical normal vector determined in a three-dimensional digital brain model to the robotic arm's base coordinate system to obtain the first target position and axis, and controlling the robotic arm to align the stimulation coil. Real-time acquisition of head pose change data of the stimulated subject is used to calculate the real-time updated coordinates of the target point in the robotic arm's base coordinate system, thereby determining the second target position and axis that the stimulation coil needs to follow. Combining the geometric and electromagnetic characteristic parameters of the stimulation coil, the tracking target pose in the robotic arm's end-effector coordinate system is calculated. Joint control commands are generated based on the deviation between the target pose and the actual pose to drive the robotic arm movement, thereby stably maintaining the effective stimulation field of the stimulation coil continuously covering the predetermined intracranial target point during head pose changes, solving the problem of off-target movement caused by head motion.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical devices, in particular to a mechanical arm dynamic tracking and follow-up positioning method and device for brain stimulation. BACKGROUND

[0002] Transcranial magnetic stimulation (TMS) is a non-invasive brain function regulation technology, which is increasingly widely used in the treatment of neuropsychiatric diseases and brain science research. The core premise of its therapeutic effect is to accurately and stably focus the time-varying magnetic field generated by the stimulation coil on the target brain function area. The traditional positioning and treatment process mainly relies on the manual operation and experience judgment of doctors: the stimulation threshold is determined by repeatedly trying and error to find the motor hotspot, and the coil is roughly placed according to the skull anatomical markers. However, this method has significant limitations: the patient is required to strictly keep the head still during the whole treatment, and any unconscious micro-movement will cause the stimulation target to deviate, resulting in "off-target", which significantly affects the certainty and repeatability of the therapeutic effect; the angle of the coil placement depends on general rules and cannot be individualized to adapt to the cortical curved surface geometry of specific brain areas, affecting the stimulation efficiency; the whole process is time-consuming and laborious, and lacks real-time feedback and verification of the actual distribution of the stimulation electric field. SUMMARY

[0003] The main purpose of the present application is to provide a mechanical arm dynamic tracking and follow-up positioning method and device for brain stimulation, which constructs a closed loop of space mapping, dynamic target point calculation, hierarchical follow-up control and physical field compensation, so that the stimulation coil carried by the mechanical arm can still dynamically, accurately and stably maintain the continuous coverage of its effective stimulation field on the predetermined intracranial target point during the free pose change of the patient's head, thereby fundamentally solving the off-target problem caused by the patient's head movement in the traditional brain stimulation treatment.

[0004] To achieve the above purpose, the present application provides a mechanical arm dynamic tracking and follow-up positioning method for brain stimulation, comprising the following steps:

[0005] Establishing a space mapping relationship between a three-dimensional digital brain model coordinate system, a head coordinate system of a stimulated object, a mechanical arm base coordinate system and a mechanical arm end tool coordinate system;

[0006] Based on the space mapping relationship, converting the stimulation target point coordinates and the initial cortical normal vector at the stimulation target point, which are determined in advance in the three-dimensional digital brain model, to the mechanical arm base coordinate system, to obtain a first target position and a first target axis;

[0007] Controlling the movement of the mechanical arm to move the stimulation coil carried by the mechanical arm end to the first target position, and aligning the axis of the stimulation coil with the first target axis;

[0008] Real-time acquisition of the pose change data of the stimulated object's head, and dynamic calculation of the real-time updated coordinates of the stimulation target point in the robotic arm's base coordinate system based on the pose change data and the spatial mapping relationship.

[0009] Based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector, the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system are determined.

[0010] Based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve;

[0011] Based on the deviation between the tracking target pose and the current actual pose of the robotic arm end-effector coordinate system, joint control commands are generated to drive the robotic arm to move, dynamically maintaining the effective stimulation field generated by the stimulation coil covering the stimulation target.

[0012] Furthermore, the steps for establishing the spatial mapping relationship between the three-dimensional digital brain model coordinate system, the head coordinate system of the stimulated object, the base coordinate system of the robotic arm, and the end effector coordinate system of the robotic arm include:

[0013] By tracking the positioning markers fixed to the head of the stimulated object using a spatial positioning system, a first real-time transformation relationship is established between the coordinate system of the head of the stimulated object and the coordinate system of the spatial positioning system.

[0014] Through calibration, a fixed tool transformation relationship is established between the coil coordinate system of the stimulation coil fixed at the end of the robotic arm and the tool coordinate system at the end of the robotic arm, and the forward kinematics relationship between the tool coordinate system at the end of the robotic arm and the base coordinate system of the robotic arm is obtained;

[0015] The surface contour of the three-dimensional digital brain model is registered with the actual surface contour of the head of the stimulated object, and the registration transformation matrix between the coordinate system of the three-dimensional digital brain model and the coordinate system of the head of the stimulated object is calculated. The three-dimensional digital brain model is a three-dimensional model reconstructed based on the head image data of the stimulated object and incorporating functional partition information of standard brain atlas.

[0016] Based on the first real-time transformation relationship, the registration transformation matrix, the fixed tool transformation relationship, and the kinematic forward solution relationship, a spatial mapping relationship is formed by linking them.

[0017] Further, based on the spatial mapping relationship, the step of transforming the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are pre-determined in the three-dimensional digital brain model, to the robotic arm base coordinate system to obtain the first target position and the first target axis includes:

[0018] In the three-dimensional digital brain model, points within the target brain functional area are selected as stimulation targets based on the fused standard brain map functional zoning information.

[0019] Based on the local surface geometry of the cortex where the stimulation target is located, the initial cortical normal vector is calculated.

[0020] The coordinates of the stimulation target point and the initial cortical normal vector are sequentially transformed into the robot arm base coordinate system through the registration transformation matrix, the first real-time transformation relationship and the kinematic forward kinematics relationship, so as to obtain the first target position and the first target axis of the coil coordinate system of the stimulation coil in the robot arm base coordinate system.

[0021] Further, the step of controlling the movement of the robotic arm to move the stimulation coil mounted at the end of the robotic arm to the first target position and aligning the axis of the stimulation coil with the axis of the first target includes:

[0022] Based on the first target position and the first target axis, the initial positioning joint angles of each joint of the robotic arm are obtained through inverse kinematics calculation;

[0023] The robotic arm is driven to move each joint to the initial positioning joint angle, so that the origin of the coil coordinate system of the stimulation coil is located at the first target position, and the axial direction of the stimulation coil is parallel to the first target axis.

[0024] Further, the step of acquiring real-time pose change data of the stimulated object's head, and dynamically calculating the real-time updated coordinates of the stimulation target point in the robotic arm's base coordinate system based on the pose change data and the spatial mapping relationship, includes:

[0025] The spatial positioning system continuously acquires real-time data of the positioning markers and updates the first real-time transformation relationship.

[0026] Using the updated first real-time transformation relationship and the registration transformation matrix, the coordinates of the stimulation target point are transformed from the three-dimensional digital brain model coordinate system to the spatial positioning system coordinate system, and then transformed to the robotic arm base coordinate system through the kinematic forward kinematics relationship chain to obtain the real-time updated coordinates of the stimulation target point.

[0027] Further, the step of determining the position of the second target and the second target axis that the stimulation coil needs to follow in the robotic arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector includes:

[0028] Based on the real-time updated coordinates and the updated first real-time transformation relationship and the registration transformation matrix, the coordinates are back-mapped to the three-dimensional digital brain model coordinate system to obtain the real-time updated cortical normal vector calculated based on the updated local surface geometric features at the position corresponding to the real-time updated coordinates.

[0029] The real-time updated coordinates and the real-time updated cortical normal vector are transformed into the robot arm base coordinate system through the spatial mapping relationship to obtain the position of the second target and the axis of the second target.

[0030] Further, the step of calculating the target pose to be achieved by the robotic arm end-effector coordinate system based on the second target position, the second target axis, and the geometric and electromagnetic field characteristic parameters of the stimulation coil includes:

[0031] Based on the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, a preset spatial offset of the origin of the coil coordinate system of the stimulation coil relative to the second target position, and a preset angular relationship between the axis of the stimulation coil and the axis of the second target are determined.

[0032] By combining the second target position, the second target axis, the preset spatial offset, and the preset angle relationship, the target coil pose of the stimulation coil in the robot arm base coordinate system is calculated.

[0033] Based on the fixed tool transformation relationship, the target coil pose is converted into the tracking target pose in the end-effector coordinate system of the robotic arm.

[0034] Further, the step of generating joint control commands based on the deviation between the tracking target pose and the current actual pose of the robotic arm end-effector coordinate system includes:

[0035] Using the target pose as the end-effector pose constraint, and combining the link parameters and joint range of motion of the robotic arm, one or more candidate joint angle combinations are obtained through inverse kinematics calculation.

[0036] Based on the optimization criteria of minimizing joint motion smoothness or energy consumption, a set of candidate joint angle combinations is selected as the target joint angle.

[0037] Based on the target joint angle and the current joint angle of the robotic arm, joint control commands are generated to drive the servo motors of each joint.

[0038] Furthermore, the step of driving the robotic arm to move and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target includes:

[0039] When the positional or orientation deviation of the deviation exceeds a first preset threshold, control all joints of the robotic arm to move in coordination.

[0040] When the positional deviation or posture deviation of the deviation is less than or equal to the first preset threshold but greater than the second preset threshold, control at least one joint of the robotic arm closest to the end to move.

[0041] The present invention also provides a robotic arm dynamic tracking and follow-up positioning device for brain stimulation, comprising:

[0042] The spatial mapping establishment unit is used to establish the spatial mapping relationship between the coordinate system of the three-dimensional digital brain model, the coordinate system of the head of the stimulated object, the base coordinate system of the robotic arm, and the coordinate system of the end tool of the robotic arm.

[0043] The initial target conversion unit is used to convert the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are determined in advance in the three-dimensional digital brain model, to the base coordinate system of the robotic arm based on the spatial mapping relationship, so as to obtain the first target position and the first target axis.

[0044] An initial positioning control unit is used to control the movement of the robotic arm, move the stimulation coil mounted on the end of the robotic arm to the first target position, and align the axis of the stimulation coil with the axis of the first target.

[0045] The target coordinate update unit is used to acquire the pose change data of the head of the stimulated object in real time, and dynamically calculate the real-time updated coordinates of the stimulation target in the base coordinate system of the robotic arm based on the pose change data and the spatial mapping relationship.

[0046] The target determination unit is used to determine the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector.

[0047] The end-effector target calculation unit is used to calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil.

[0048] The follow-up control and drive unit is used to generate joint control commands based on the deviation between the pose of the tracking target and the current actual pose of the end-effector coordinate system of the robotic arm, drive the robotic arm to move, and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target.

[0049] The present invention provides a method and device for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation, which has the following beneficial effects: By fusing individual images with standard brain maps, the present invention achieves visualized and precise selection of functional target points in a three-dimensional digital brain model, and calculates the optimal stimulation direction (normal vector) based on local cortical geometry dynamics, solving the problems of coarse positioning and experience-dependent angles in traditional methods. Through high-frequency head tracking and real-time coordinate updates, the coordinates of the moving physical target point are calculated in real time to the robot coordinate system, and a hierarchical control strategy (multi-joint coordination for large deviations, and fine-tuning of the end joints for small deviations) is adopted to achieve millisecond-level dynamic compensation for unconscious head movements of the patient, fundamentally solving the "off-target" problem in the treatment process and ensuring that the effective stimulation field continuously covers the target point. Furthermore, by integrating the electromagnetic physical parameters of the stimulation coil for feedforward compensation and incorporating six-dimensional force feedback into the control loop, the safety of operation and the smoothness of human-machine interaction are improved while ensuring sub-millimeter-level follow-up accuracy. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating a method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to an embodiment of the present invention.

[0051] Figure 2 This is a structural block diagram of a robotic arm dynamic tracking and follow-up positioning device for brain stimulation according to an embodiment of the present invention.

[0052] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0054] Reference Figure 1 This is a flowchart illustrating a method for dynamic tracking and follow-up localization of a robotic arm for brain stimulation proposed in this invention, comprising the following steps:

[0055] S1. Establish the spatial mapping relationship between the coordinate system of the three-dimensional digital brain model, the coordinate system of the head of the stimulated object, the base coordinate system of the robotic arm, and the coordinate system of the end tool of the robotic arm.

[0056] S2, based on the spatial mapping relationship, the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are determined in advance in the three-dimensional digital brain model, are transformed to the base coordinate system of the robotic arm to obtain the first target position and the first target axis;

[0057] S3, control the movement of the robotic arm to move the stimulation coil mounted at the end of the robotic arm to the first target position, and align the axis of the stimulation coil with the axis of the first target.

[0058] S4. Real-time acquisition of the pose change data of the head of the stimulated object, and dynamic calculation of the real-time updated coordinates of the stimulation target point in the base coordinate system of the robotic arm based on the pose change data and the spatial mapping relationship.

[0059] S5, based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector, determine the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system;

[0060] S6. Based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve.

[0061] S7. Based on the deviation between the tracking target pose and the current actual pose of the end-effector coordinate system of the robotic arm, a joint control command is generated to drive the robotic arm to move and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target.

[0062] In one embodiment, for step S1,

[0063] The steps for establishing the spatial mapping relationship between the three-dimensional digital brain model coordinate system, the head coordinate system of the stimulated object, the base coordinate system of the robotic arm, and the end effector coordinate system of the robotic arm include:

[0064] By tracking the positioning markers fixed to the head of the stimulated object using a spatial positioning system, a first real-time transformation relationship is established between the coordinate system of the head of the stimulated object and the coordinate system of the spatial positioning system.

[0065] Through calibration, a fixed tool transformation relationship is established between the coil coordinate system of the stimulation coil fixed at the end of the robotic arm and the tool coordinate system at the end of the robotic arm, and the forward kinematics relationship between the tool coordinate system at the end of the robotic arm and the base coordinate system of the robotic arm is obtained;

[0066] The surface contour of the three-dimensional digital brain model is registered with the actual surface contour of the head of the stimulated object, and the registration transformation matrix between the coordinate system of the three-dimensional digital brain model and the coordinate system of the head of the stimulated object is calculated. The three-dimensional digital brain model is a three-dimensional model reconstructed based on the head image data of the stimulated object and incorporating functional partition information of standard brain atlas.

[0067] Based on the first real-time transformation relationship, the registration transformation matrix, the fixed tool transformation relationship, and the kinematic forward solution relationship, a spatial mapping relationship is formed by linking them.

[0068] In practice, a high-precision optical (or electromagnetic) spatial positioning system continuously tracks several infrared reflective markers or electromagnetic sensors (i.e., positioning markers) fixed to a specialized positioning headgear worn by the subject. These markers define a head coordinate system that changes with head movement. The spatial positioning system measures the pose (position and orientation) of this head coordinate system relative to its own fixed spatial positioning system coordinate system in real time, establishing a first real-time transformation relationship between the two (e.g., a 4x4 homogeneous transformation matrix). This relationship is dynamic and reflects any head movement or rotation in real time. The robotic arm system is pre-calibrated. Through a calibration process (e.g., using the standard tool tip method or a three-dimensional calibration block), the relative position and orientation relationship between the coil coordinate system of the stimulation coil (defined with its electromagnetic or geometric center as the origin and its principal axis as the Z-axis) and the tool coordinate system at the end of the robotic arm (fixed to the end flange of the robotic arm) is determined. This is the fixed tool transformation relationship, which remains unchanged throughout a single treatment session. Based on the DH parameters of the robotic arm, the forward kinematics relationship from the end-effector coordinate system to the base coordinate system (a reference system fixed to the robotic arm base) is obtained. This relationship is uniquely determined by the real-time angle values ​​of each joint of the robotic arm. A crucial registration operation is then performed. The 3D digital brain model used here is a high-precision 3D model reconstructed from medical imaging data such as MRI of the individual being stimulated. It incorporates functional partitioning information from standard brain atlases such as AAL and Brodmann using algorithms. The model not only possesses anatomical structure but also carries functional location information. Through point cloud registration, surface matching, and other algorithms, the surface contour of the digital brain model is precisely aligned with the actual surface contour of the subject's head obtained through a spatial positioning system or the subject's head. The registration transformation matrix between the 3D digital brain model coordinate system and the subject's head coordinate system is calculated, thereby binding the virtual brain anatomy and functional atlas to the real physical head. The four core transformation relationships mentioned above are mathematically chained together or synthesized: starting from the coordinate system of the three-dimensional digital brain model, they are sequentially processed through the registration transformation matrix (linked to the head coordinate system), the first real-time transformation relationship (linked to the spatial positioning system coordinate system, and linked back to the robotic arm base coordinate system through inverse transformation or direct derivation), the forward kinematics relationship (linked to the end-effector coordinate system), and the fixed-tool transformation relationship (linked to the coil coordinate system), forming a complete and reversibly derivable spatial mapping relationship. Through this mapping relationship, the coordinates of any point in the brain model (such as the stimulation target point) can be accurately transformed to the robotic arm base coordinate system to guide the movement of the robotic arm; conversely, the actual pose of the robotic arm end-effector or coil can also be mapped back to the brain model for visualization and verification, providing a unified spatiotemporal reference for the entire dynamic tracking process.

[0069] In one embodiment, for step S2,

[0070] Based on the spatial mapping relationship, the step of transforming the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are pre-determined in the three-dimensional digital brain model, to the base coordinate system of the robotic arm to obtain the position of the first target and the axis of the first target includes:

[0071] In the three-dimensional digital brain model, points within the target brain functional area are selected as stimulation targets based on the fused standard brain map functional zoning information.

[0072] Based on the local surface geometry of the cortex where the stimulation target is located, the initial cortical normal vector is calculated.

[0073] The coordinates of the stimulation target point and the initial cortical normal vector are sequentially transformed into the robot arm base coordinate system through the registration transformation matrix, the first real-time transformation relationship and the kinematic forward kinematics relationship, so as to obtain the first target position and the first target axis of the coil coordinate system of the stimulation coil in the robot arm base coordinate system.

[0074] In practice, step S2 involves accurately translating the clinical treatment intent (which brain region to stimulate and at what angle) from the virtual digital space into executable pose commands in the physical robot space. This is based on target localization within a 3D digital brain model: using standard brain atlas functional zoning information (e.g., Brodmann areas, AAL templates) that has been precisely integrated with the individual brain model, specific points located within the target brain functional area (e.g., M1 area of ​​the primary motor cortex, DLPFC area of ​​the dorsolateral prefrontal cortex) are directly selected on the visualized 3D model or automatically recommended by the system as stimulation target points. After determining the spatial coordinates of the target point, to further optimize the biological effect of the stimulation, the initial cortical normal vector is calculated. This vector is based on the local geometric features of the cortical surface at the target point's location: by extracting the vertices of the cortical surface within a certain range around the target point, principal component analysis or surface fitting algorithms are used to calculate the normal vector perpendicular to this local cortical surface. This vector direction is physiologically considered the optimal stimulation direction because it maximizes the perpendicular penetration of the magnetic field generated by the stimulation coil through the skull and effectively induces an induced electric field in cortical neurons. Therefore, the target point (target coordinates) and optimal orientation (initial cortical normal vector) in virtual space are determined. Using the spatial mapping relationship established in step S1, these two parameters are transmitted to the real world. Specifically, the target coordinates and the initial cortical normal vector (as a direction vector) are transformed from the 3D digital brain model coordinate system to the head coordinate system of the stimulated object through a registration transformation matrix; combined with the first real-time transformation relationship at this moment (initial time), they are further transformed to the spatial positioning system coordinate system; through the system's kinematic forward kinematics relationship, they are transformed to the robot arm base coordinate system. After this series of rigid body transformations, the first target position (i.e., the spatial coordinates of the electromagnetic center of the stimulation coil to reach in order to stimulate the target point) and the first target axis (i.e., the direction vector that the principal axis of the stimulation coil should align with in order to obtain the optimal stimulation angle) described in the robot arm base coordinate system are obtained. These constitute the absolute spatial command for driving the robot arm to perform initial positioning, completing the precise mapping from the target point in the brain to the robot's end-effector pose.

[0075] In one embodiment, for step S3,

[0076] The step of controlling the movement of the robotic arm to move the stimulation coil mounted at the end of the robotic arm to the first target position and aligning the axis of the stimulation coil with the axis of the first target includes:

[0077] Based on the first target position and the first target axis, the initial positioning joint angles of each joint of the robotic arm are obtained through inverse kinematics calculation;

[0078] The robotic arm is driven to move each joint to the initial positioning joint angle, so that the origin of the coil coordinate system of the stimulation coil is located at the first target position, and the axial direction of the stimulation coil is parallel to the first target axis.

[0079] In practical implementation, the first target position and the first target axis are used as pose constraints for the end effector of the robotic arm (which, through a fixed tool transformation relationship, corresponds to the stimulation coil). Since the control of the robotic arm ultimately acts on its various rotational or translational joints, it is necessary to convert the spatial pose requirements of the end effector into the angles that each joint needs to rotate. This process is completed through inverse kinematics calculation. The inverse kinematics algorithm solves for one or more sets of joint angle combinations that enable the coordinate system of the end effector to reach a specified position and orientation based on the geometric model of the robotic arm (such as DH parameters) and the kinematic chain. Based on criteria such as avoiding singular configurations, minimizing the range of joint motion, or optimizing energy, a unique set is selected from multiple feasible solutions as the initial positioning joint angles. This set of angle values ​​is the direct position command driving the servo motors of each joint. The control system sends this set of initial positioning joint angles as target values ​​to the servo drivers of each joint of the robotic arm. Under closed-loop position control, each joint moves smoothly and synchronously to its specified angle. When all joints reach their target angles, according to the forward kinematics principle of the robotic arm, its end-effector coordinate system will inevitably reach the unique spatial pose obtained by inverse kinematics. Through the previously calibrated fixed-tool transformation relationship, the actual spatial pose of the stimulation coil's coordinate system can be determined. The final verification goal of step S3 is to achieve dual composite alignment: first, positional alignment, meaning the origin of the stimulation coil's coordinate system coincides spatially with the first target position; second, axial alignment, meaning the axial direction of the stimulation coil (usually defined as the Z-axis direction of the coil coordinate system) remains parallel to the first target axis. This ensures that at the start of treatment, the coil's center point is aligned with the functional target point in the brain, and the direction of its generated magnetic field is consistent with the optimal stimulation direction calculated based on individual cortical geometry, laying a precise physical foundation for subsequent treatment. The entire initial positioning process can be automatically completed under system monitoring, replacing the time-consuming and error-prone manual positioning operation that relies on the doctor's hand-eye coordination in traditional treatment.

[0080] In one embodiment, for step S4,

[0081] The steps of acquiring real-time pose change data of the stimulated object's head, and dynamically calculating the real-time updated coordinates of the stimulation target point in the robotic arm's base coordinate system based on the pose change data and the spatial mapping relationship, include:

[0082] The spatial positioning system continuously acquires real-time data of the positioning markers and updates the first real-time transformation relationship.

[0083] Using the updated first real-time transformation relationship and the registration transformation matrix, the coordinates of the stimulation target point are transformed from the three-dimensional digital brain model coordinate system to the spatial positioning system coordinate system, and then transformed to the robotic arm base coordinate system through the kinematic forward kinematics relationship chain to obtain the real-time updated coordinates of the stimulation target point.

[0084] In practice, a spatial positioning system (such as an infrared optical camera array) continuously captures the spatial coordinates of positioning markers fixed on the head positioning device at frequencies of tens or even hundreds of hertz, enabling continuous high-frequency monitoring of head movements. Whenever a new frame of marker data is acquired, the current pose of the stimulated object's head coordinate system relative to the spatial positioning system coordinate system, defined by these markers, is immediately recalculated. Based on this, the first real-time transformation relationship established in step S1 is updated. The transformation relationship thus transforms from a static parameter into a dynamic function that varies with time, encoding the displacement and rotation of the head at every moment.

[0085] Using this updated first real-time transformation relation, the absolute coordinates of the stimulation target in space are recalculated. Although the coordinates of the stimulation target in the 3D digital brain model coordinate system are fixed (predetermined in step S2), the corresponding position of these fixed coordinates in the physical world (e.g., the robotic arm base coordinate system) has changed due to head movement. To find this new position, the spatial mapping relation chain constructed in step S1 needs to be reactivated. The specific path is as follows: the fixed stimulation target coordinates are transformed from the 3D digital brain model coordinate system to the head coordinate system of the stimulated object that follows head movement through an invariant registration transformation matrix; the updated first real-time transformation relation is immediately applied to transform it to the spatial positioning system coordinate system; and then, through the same fixed kinematic forward kinematic relation, it is mapped to the robotic arm base coordinate system. By performing this series of continuous coordinate transformations, a real-time updated coordinate value defined in the robotic arm base coordinate system is finally output. This real-time updated coordinate refers to the actual physical position of the preset brain stimulation target in the robot coordinate system at the current moment, when the head is in the current pose, providing a stable and accurate spatial target for the follow-up control of the robotic arm.

[0086] In one embodiment, for step S5,

[0087] The step of determining the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robotic arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector includes:

[0088] Based on the real-time updated coordinates and the updated first real-time transformation relationship and the registration transformation matrix, the coordinates are back-mapped to the three-dimensional digital brain model coordinate system to obtain the real-time updated cortical normal vector calculated based on the updated local surface geometric features at the position corresponding to the real-time updated coordinates.

[0089] The real-time updated coordinates and the real-time updated cortical normal vector are transformed into the robot arm base coordinate system through the spatial mapping relationship to obtain the position of the second target and the axis of the second target.

[0090] In practice, head movements can cause subtle but significant changes in the relative spatial relationship between the scalp surface and the underlying cortex. Therefore, simply using the initial cortical normal vector calculated based on the static head pose to guide the coil angles during dynamic processes may no longer be optimal. To address this issue, this step performs a reverse spatial mapping and real-time geometric calculation. Specifically, using the real-time updated coordinates obtained in step S4 (which are essentially the physical position of the target point in the robotic arm's base coordinate system under the current head pose), combined with the updated first real-time transformation relationship and the fixed registration transformation matrix, the anatomical position of the target point in the virtual space of the 3D digital brain model under the current physical head pose is accurately calculated through the inverse transformation chain. Locally, surface geometric analysis (such as local surface fitting) is re-performed on the cortical surface around this "real-time mapped point" to calculate the real-time updated cortical normal vector corresponding to the instantaneous head pose and cortical curvature. This process ensures that the normal vector of the guide coil angle is always perpendicular to the actual cortical anatomy and geometry at the target point under the scalp at the current moment, achieving synchronous adaptive optimization of the stimulation angle and head posture changes. After obtaining the real-time updated coordinates representing "where the target is" and the real-time updated cortical normal vector representing "what angle is optimal," the ultimate goal of step S5 is to transform this pair of information back into the controllable space of the robotic arm: calling the complete spatial mapping relationship chain, transforming this pair of parameters from the three-dimensional digital brain model coordinate system, through the transformation relationship corresponding to the current head posture, and finally to the robotic arm base coordinate system. After this transformation, the second target position and the second target axis are obtained. The second target position indicates the new physical position that the center of the stimulation coil needs to move to in order to align with the original target point in the brain under the current head posture; the second target axis indicates the new direction that the axis of the stimulation coil should point to at the current moment in order to maintain the optimal perpendicular relationship with the cortex. Through the refreshed "position-axis" command, precise and adaptive target input is provided for the real-time motion control of the robotic arm, thereby ensuring that the spatial target point of stimulation and the energy delivery direction can be dynamically kept accurate and optimal throughout the treatment process.

[0091] In one embodiment, for step S6,

[0092] The steps for calculating the target pose to be achieved by the robotic arm end-effector coordinate system based on the second target position, the second target axis, and the geometric and electromagnetic field characteristics of the stimulation coil include:

[0093] Based on the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, a preset spatial offset of the origin of the coil coordinate system of the stimulation coil relative to the second target position, and a preset angular relationship between the axis of the stimulation coil and the axis of the second target are determined.

[0094] By combining the second target position, the second target axis, the preset spatial offset, and the preset angle relationship, the target coil pose of the stimulation coil in the robot arm base coordinate system is calculated.

[0095] Based on the fixed tool transformation relationship, the target coil pose is converted into the tracking target pose in the end-effector coordinate system of the robotic arm.

[0096] In practical implementation, the second target position and second target axis determined by step S5 describe the position and orientation of the ideal "stimulation point" to optimize the biological effect of stimulation. However, the actual physical stimulation coil, as an entity with a specific three-dimensional shape and electromagnetic field spatial distribution characteristics, usually does not have its maximum or most effective stimulation electric field region (i.e., the effective stimulation field) strictly located at the geometric center of the coil, and its optimal stimulation direction may also have a fixed angle relationship with the mechanical axis of the coil. Therefore, step S6 needs to compensate and transform the ideal target according to the specific physical characteristics of the coil to calculate the exact pose required to drive the end effector of the robotic arm. Specifically, this process relies on prior knowledge or calibration results of the geometric parameters (such as coil shape, size, winding structure) and electromagnetic field characteristic parameters (such as the electric field distribution model obtained through electromagnetic simulation or measurement) of the stimulation coil. Based on these parameters, two key compensation quantities are predetermined: one is a preset spatial offset, which defines the relative position vector that needs to be maintained between the origin of the coil coordinate system of the stimulation coil (e.g., the geometric installation center) and the second target position (ideal stimulation point). This offset ensures that once the coil is positioned as instructed, the center of its effective stimulation field coincides with the target point in the brain. Secondly, a preset angular relationship defines the fixed angular difference that needs to be maintained between the axis of the stimulation coil (i.e., its principal normal direction) and the axis of the second target (the ideal stimulation direction) (for example, the optimal stimulation direction of some figure-eight coils is perpendicular to the coil plane). These two compensation amounts concretize the abstract physiological stimulation requirements into precise constraints on the physical coil pose. Calculations are performed: In the robotic arm's base coordinate system, using the second target position as a reference point, a preset spatial offset is superimposed to obtain the actual spatial point where the origin of the stimulation coil's coordinate system should be located; simultaneously, the second target axis is rotated according to the preset angular relationship to obtain the actual direction in which the axis of the stimulation coil should point. Combining the corrected position and direction yields the target coil pose that the stimulation coil's coordinate system should achieve in the robotic arm's base coordinate system. Since the robotic arm directly controls its end flange, not the coil itself, a final coordinate transformation is required. The known fixed tool transformation relationship obtained through calibration in step S1 is invoked to perform an inverse transformation on the calculated target coil pose, thereby solving for the corresponding pose that the end-effector coordinate system of the robotic arm needs to achieve in order to reach the coil pose, i.e., the tracking target pose. This tracking target pose is the final spatial instruction that the robotic arm motion controller can directly understand and execute. Through the precise compensation and transformation in step S6, it is ensured that the ultimate goal of each motion control of the robotic arm is to physically achieve continuous and precise coverage of the target point in the brain during movement by an effective stimulus field, rather than simply mechanically following a geometric point, thereby improving the accuracy of dynamic tracking from "mechanical alignment" to the level of "biological effect alignment".

[0097] In one embodiment, for step S7,

[0098] The step of generating joint control commands based on the deviation between the tracking target pose and the current actual pose of the robotic arm end-effector coordinate system includes:

[0099] Using the target pose as the end-effector pose constraint, and combining the link parameters and joint range of motion of the robotic arm, one or more candidate joint angle combinations are obtained through inverse kinematics calculation.

[0100] Based on the optimization criteria of minimizing joint motion smoothness or energy consumption, a set of candidate joint angle combinations is selected as the target joint angle.

[0101] Based on the target joint angle and the current joint angle of the robotic arm, joint control commands are generated to drive the servo motors of each joint.

[0102] In practical implementation, the calculated target pose is used as a hard constraint on the robotic arm's end effector in Cartesian space (position and orientation). To satisfy this constraint, each joint of the robotic arm must move to a specific set of angles. The inverse kinematics solver is invoked, and based on the precise link parameters (such as length and torsion angle) of the robotic arm, and considering the physical range of motion of each joint, all mathematically feasible solutions that allow the end effector coordinate system to reach the target pose are solved, i.e., one or more combinations of candidate joint angles. Due to the characteristics of redundant degree-of-freedom robotic arms (such as six degrees of freedom and above), the same end effector pose often corresponds to multiple different joint configurations. Intelligent screening is performed based on preset optimization criteria, such as joint motion smoothness (i.e., selecting the solution with the smallest sum of joint angle changes compared to the previous control cycle, reducing frequent motor starts and stops and mechanical vibration) or lowest energy consumption (i.e., prioritizing solutions that keep each joint away from its extreme position and in its efficient working range). By applying the optimization criteria, a set of solutions is selected from all candidate solutions as the final target joint angles for the current control cycle. By optimizing the selection process, the robotic arm's motion is ensured to be energy-efficient, smooth, and continuous when tracking rapidly changing head movements, avoiding severe shaking and unusual movements that could cause patient discomfort or affect system lifespan. The selected target joint angle is compared with the current joint angle of the robotic arm, fed back in real-time by the encoder, to calculate the angle error of each joint. Based on this error, servo control algorithms such as proportional-integral-derivative (PI-DI) control are used to generate joint control commands (usually torque or speed commands) that are sent to the servo motors of each joint in real time. These commands drive the motors to coordinate their movements, causing the joint angles of the robotic arm to converge towards the target angle, thereby moving the end effector's stimulation coil toward the target pose.

[0103] In one embodiment, the step of driving the robotic arm to move and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target includes:

[0104] When the positional or orientation deviation of the deviation exceeds a first preset threshold, control all joints of the robotic arm to move in coordination.

[0105] When the positional deviation or posture deviation of the deviation is less than or equal to the first preset threshold but greater than the second preset threshold, control at least one joint of the robotic arm closest to the end to move.

[0106] In practice, the process of driving the robotic arm to dynamically maintain an effective stimulation field covering the target point does not employ a single, fixed control mode, but rather an adaptive hierarchical control strategy. The core of this strategy lies in intelligently allocating the robotic arm's motion resources based on the magnitude and nature of the deviation between the target's pose and the actual pose of the robotic arm's end effector. This ensures tracking accuracy and real-time performance while optimizing motion smoothness, reducing overall energy consumption, and minimizing psychological disturbance to the patient caused by large robotic arm movements. Specifically, the deviations constituted by position and orientation errors are continuously monitored. When a large-amplitude or rapid head movement causes either the position or posture deviation to exceed a first preset threshold, a rapid, large-scale pose correction is deemed necessary. At this point, all joints of the robotic arm are controlled to move in a coordinated manner. All degrees of freedom are mobilized to enable the robotic arm to respond to large target changes in the most direct and fastest path, ensuring that the stimulation coil can quickly keep up with the head's displacement and preventing "missing the target" due to response delay. This mode is similar to "macro-motion" or "coarse tracking," prioritizing the tracking capture range and response speed. When the deviation is reduced through the aforementioned coordinated movement, bringing both positional and posture deviations down to within the first preset threshold but still exceeding the second preset threshold (the second preset threshold is less than the first preset threshold), it indicates that head movement has transitioned into small-amplitude micro-movements or tremors (such as adjustments caused by breathing or slight discomfort). At this point, continuing to drive all joints in response would not only be inefficient and energy-intensive but could also introduce unnecessary vibrations due to the linkage of multiple joints. Therefore, switching to fine-tuning mode primarily controls the movement of at least one joint closest to the end effector (such as the last two or three joints). Since the end effector joint is closest to the stimulation coil, its small-range movements have the most sensitive and direct impact on the end effector pose, enabling compensation for residual deviations with extremely high resolution. This mode is similar to "micro-movement" or "precise tracking," fully utilizing the dynamic characteristics of the robotic arm to achieve high-bandwidth, high-stability fine-tuning within a very small range, thereby effectively filtering out the influence of high-frequency head tremors on stimulation localization. Through this two-level adaptive control based on deviation magnitude, dynamic follow-up performance is optimized: rapid reset when large deviations occur, and precise stabilization during minor disturbances. This not only efficiently maintains the continuous coverage of the stimulation target by the effective stimulation field, but also significantly improves the smoothness and reliability of the entire treatment process, enhancing patient comfort and treatment experience.

[0107] Reference Figure 2 The diagram shows a structural block diagram of a robotic arm dynamic tracking and follow-up positioning device for brain stimulation according to an embodiment of the present invention, comprising:

[0108] The spatial mapping establishment unit is used to establish the spatial mapping relationship between the coordinate system of the three-dimensional digital brain model, the coordinate system of the head of the stimulated object, the base coordinate system of the robotic arm, and the coordinate system of the end tool of the robotic arm.

[0109] The initial target conversion unit is used to convert the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are determined in advance in the three-dimensional digital brain model, to the base coordinate system of the robotic arm based on the spatial mapping relationship, so as to obtain the first target position and the first target axis.

[0110] An initial positioning control unit is used to control the movement of the robotic arm, move the stimulation coil mounted on the end of the robotic arm to the first target position, and align the axis of the stimulation coil with the axis of the first target.

[0111] The target coordinate update unit is used to acquire the pose change data of the head of the stimulated object in real time, and dynamically calculate the real-time updated coordinates of the stimulation target in the base coordinate system of the robotic arm based on the pose change data and the spatial mapping relationship.

[0112] The target determination unit is used to determine the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector.

[0113] The end-effector calculation unit is used to calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil.

[0114] The follow-up control and drive unit is used to generate joint control commands based on the deviation between the tracking target pose and the current actual pose of the end-effector coordinate system of the robotic arm, drive the robotic arm to move, and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target.

[0115] For the specific implementation of each unit in the above device example, please refer to the method embodiments described above, and will not be repeated here.

[0116] In summary, this invention establishes a spatial mapping relationship between multiple coordinate systems, transforming the coordinates of the stimulation target point and the initial cortical normal vector determined in the three-dimensional digital brain model to the robotic arm's base coordinate system. This yields the first target position and axis, and the robotic arm is controlled to align the stimulation coil. Real-time acquisition of head pose change data of the stimulated subject is used to dynamically calculate the real-time updated coordinates of the target point in the robotic arm's base coordinate system, thereby determining the second target position and axis that the stimulation coil needs to follow. Combining the geometric and electromagnetic characteristics of the stimulation coil, the tracking target pose in the robotic arm's end-effector coordinate system is calculated. Joint control commands are generated based on the deviation between this target pose and the actual pose, driving the robotic arm to move. This dynamically, accurately, and stably maintains the effective stimulation field of the stimulation coil continuously covering the predetermined intracranial target point during free head pose changes, fundamentally solving the problem of off-target movement caused by head motion.

[0117] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for dynamic tracking and servo positioning of a robotic arm for brain stimulation, characterized in that, Includes the following steps: Establish the spatial mapping relationship between the coordinate system of the three-dimensional digital brain model, the coordinate system of the head of the stimulated object, the base coordinate system of the robotic arm, and the coordinate system of the end effector of the robotic arm; Based on the spatial mapping relationship, the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which were determined in advance in the three-dimensional digital brain model, are transformed to the base coordinate system of the robotic arm to obtain the position of the first target and the axis of the first target. Control the movement of the robotic arm to move the stimulation coil mounted at the end of the robotic arm to the first target position, and align the axis of the stimulation coil with the axis of the first target. Real-time acquisition of the pose change data of the stimulated object's head, and dynamic calculation of the real-time updated coordinates of the stimulation target point in the robotic arm's base coordinate system based on the pose change data and the spatial mapping relationship. Based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector, the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system are determined. Based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve; Based on the deviation between the tracking target pose and the current actual pose of the robotic arm end-effector coordinate system, joint control commands are generated to drive the robotic arm to move, dynamically maintaining the effective stimulation field generated by the stimulation coil covering the stimulation target.

2. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 1, characterized in that, The steps for establishing the spatial mapping relationship between the three-dimensional digital brain model coordinate system, the head coordinate system of the stimulated object, the base coordinate system of the robotic arm, and the end effector coordinate system of the robotic arm include: By tracking the positioning markers fixed to the head of the stimulated object using a spatial positioning system, a first real-time transformation relationship is established between the coordinate system of the head of the stimulated object and the coordinate system of the spatial positioning system. Through calibration, a fixed tool transformation relationship is established between the coil coordinate system of the stimulation coil fixed at the end of the robotic arm and the tool coordinate system at the end of the robotic arm, and the forward kinematics relationship between the tool coordinate system at the end of the robotic arm and the base coordinate system of the robotic arm is obtained; The surface contour of the three-dimensional digital brain model is registered with the actual surface contour of the head of the stimulated object, and the registration transformation matrix between the coordinate system of the three-dimensional digital brain model and the coordinate system of the head of the stimulated object is calculated. The three-dimensional digital brain model is a three-dimensional model reconstructed based on the head image data of the stimulated object and incorporating functional partition information of standard brain atlas. Based on the first real-time transformation relationship, the registration transformation matrix, the fixed tool transformation relationship, and the kinematic forward solution relationship, a spatial mapping relationship is formed by linking them.

3. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 2, characterized in that, The step of transforming the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, pre-determined in the three-dimensional digital brain model, to the robotic arm base coordinate system based on the spatial mapping relationship to obtain the first target position and the first target axis includes: In the three-dimensional digital brain model, points within the target brain functional area are selected as stimulation targets based on the fused standard brain map functional zoning information. Based on the local surface geometry of the cortex where the stimulation target is located, the initial cortical normal vector is calculated. The coordinates of the stimulation target point and the initial cortical normal vector are sequentially transformed into the robot arm base coordinate system through the registration transformation matrix, the first real-time transformation relationship and the kinematic forward kinematics relationship, so as to obtain the first target position and the first target axis of the coil coordinate system of the stimulation coil in the robot arm base coordinate system.

4. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 3, characterized in that, The step of controlling the movement of the robotic arm to move the stimulation coil mounted on the end of the robotic arm to the first target position and aligning the axis of the stimulation coil with the axis of the first target includes: Based on the first target position and the first target axis, the initial positioning joint angles of each joint of the robotic arm are obtained through inverse kinematics calculation; The robotic arm is driven to move each joint to the initial positioning joint angle, so that the origin of the coil coordinate system of the stimulation coil is located at the first target position, and the axial direction of the stimulation coil is parallel to the first target axis.

5. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 2, characterized in that, The step of acquiring real-time pose change data of the stimulated object's head and dynamically calculating the real-time updated coordinates of the stimulation target point in the robotic arm's base coordinate system based on the pose change data and the spatial mapping relationship includes: The spatial positioning system continuously acquires real-time data of the positioning markers and updates the first real-time transformation relationship. Using the updated first real-time transformation relationship and the registration transformation matrix, the coordinates of the stimulation target point are transformed from the three-dimensional digital brain model coordinate system to the spatial positioning system coordinate system, and then transformed to the robotic arm base coordinate system through the kinematic forward kinematics relationship chain to obtain the real-time updated coordinates of the stimulation target point.

6. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 5, characterized in that, The step of determining the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robotic arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector includes: Based on the real-time updated coordinates and the updated first real-time transformation relationship and the registration transformation matrix, the coordinates are back-mapped to the three-dimensional digital brain model coordinate system to obtain the real-time updated cortical normal vector calculated based on the updated local surface geometric features at the position corresponding to the real-time updated coordinates. The real-time updated coordinates and the real-time updated cortical normal vector are transformed into the robot arm base coordinate system through the spatial mapping relationship to obtain the position of the second target and the axis of the second target.

7. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 2, characterized in that, The step of calculating the target pose to be achieved by the robotic arm end-effector coordinate system based on the second target position, the second target axis, and the geometric and electromagnetic field characteristics of the stimulation coil includes: Based on the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil, a preset spatial offset of the origin of the coil coordinate system of the stimulation coil relative to the second target position, and a preset angular relationship between the axis of the stimulation coil and the axis of the second target are determined. By combining the second target position, the second target axis, the preset spatial offset, and the preset angle relationship, the target coil pose of the stimulation coil in the robot arm base coordinate system is calculated. Based on the fixed tool transformation relationship, the target coil pose is converted into the tracking target pose in the end-effector coordinate system of the robotic arm.

8. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 1 or 7, characterized in that, The step of generating joint control commands based on the deviation between the tracking target pose and the current actual pose of the robotic arm end-effector coordinate system includes: Using the target pose as the end-effector pose constraint, and combining the link parameters and joint range of motion of the robotic arm, one or more candidate joint angle combinations are obtained through inverse kinematics calculation. Based on the optimization criteria of minimizing joint motion smoothness or energy consumption, a set of candidate joint angle combinations is selected as the target joint angle. Based on the target joint angle and the current joint angle of the robotic arm, joint control commands are generated to drive the servo motors of each joint.

9. The method for dynamic tracking and follow-up positioning of a robotic arm for brain stimulation according to claim 8, characterized in that, The step of driving the robotic arm to move and dynamically maintaining the effective stimulation field generated by the stimulation coil covering the stimulation target includes: When the positional or orientation deviation of the deviation exceeds a first preset threshold, control all joints of the robotic arm to move in coordination. When the positional deviation or posture deviation of the deviation is less than or equal to the first preset threshold but greater than the second preset threshold, control at least one joint of the robotic arm closest to the end to move.

10. A robotic arm dynamic tracking and follow-up positioning device for brain stimulation, characterized in that, include: The spatial mapping establishment unit is used to establish the spatial mapping relationship between the coordinate system of the three-dimensional digital brain model, the coordinate system of the head of the stimulated object, the base coordinate system of the robotic arm, and the coordinate system of the end tool of the robotic arm. The initial target conversion unit is used to convert the coordinates of the stimulation target point and the initial cortical normal vector at the stimulation target point, which are determined in advance in the three-dimensional digital brain model, to the base coordinate system of the robotic arm based on the spatial mapping relationship, so as to obtain the first target position and the first target axis. An initial positioning control unit is used to control the movement of the robotic arm, move the stimulation coil mounted on the end of the robotic arm to the first target position, and align the axis of the stimulation coil with the axis of the first target. The target coordinate update unit is used to acquire the pose change data of the head of the stimulated object in real time, and dynamically calculate the real-time updated coordinates of the stimulation target in the base coordinate system of the robotic arm based on the pose change data and the spatial mapping relationship. The target determination unit is used to determine the position of the second target and the axis of the second target that the stimulation coil needs to follow in the robot arm base coordinate system based on the real-time updated coordinates and the corresponding real-time updated cortical normal vector. The end-effector target calculation unit is used to calculate the target pose that the end-effector coordinate system of the robotic arm needs to achieve based on the second target position, the second target axis, and the geometric parameters and electromagnetic field characteristic parameters of the stimulation coil. The follow-up control and drive unit is used to generate joint control commands based on the deviation between the pose of the tracking target and the current actual pose of the end-effector coordinate system of the robotic arm, drive the robotic arm to move, and dynamically maintain the effective stimulation field generated by the stimulation coil covering the stimulation target.

Images