A nano-vision micro-control robot system and a control method thereof

Abstract

The application provides a nanometer vision micro-control robot system and a control method thereof, and relates to the technical field of micro-nano operation robots.The application adopts a composite configuration of a large-stroke motion module, a multi-degree-of-freedom manipulator and a micro-positioning module, specifically, the large-stroke motion module provides macro movement ability for finding a target, the multi-degree-of-freedom manipulator provides flexible posture adjustment ability for approaching the target, and the micro-positioning module provides nanometer-level accurate positioning ability for final operation, and through the control system, the Cartesian space error of the end is mapped and distributed to each joint through inverse kinematics calculation, so that parasitic displacement caused by joint coupling of the multi-degree-of-freedom manipulator is effectively decomposed and compensated; meanwhile, through real-time tuning of PID parameters, the remaining unmodeled dynamics and disturbances can be effectively coped with, so that the whole system can still show high control precision and stability when facing multiple complex nonlinear factors such as vacuum dry friction, hysteresis and motion coupling.

Description

Technical Field

[0001] This invention relates to the field of micro-nano manipulation robot technology, specifically to a nano-view micro-controlled robot system and its control method. Background Technology

[0002] In cutting-edge fields such as microelectromechanical systems (MEMS) packaging, semiconductor failure analysis, and single-cell manipulation in life sciences, in-situ observation and fine manipulation of micro- and nano-scale target particles (such as nanoparticles, carbon nanotubes, and biological cells) are required under the high-vacuum environment of scanning electron microscopy (SEM). SEM provides nanometer-resolution observation capabilities, but achieving high-performance in-situ manipulation under this environment faces numerous technical challenges. Nanoscale micro-manipulation refers to in-situ manipulation of micro- and nano-particles in microscopic environments such as those using scanning electron microscopes.

[0003] Existing technical solutions are mostly based on micro-nano positioning platforms with three-axis orthogonal translation, which have limited degrees of freedom and are difficult to effectively "observe and manipulate" complex three-dimensional microstructures (such as the inner walls of holes and inclined surfaces). In addition, in order to balance large stroke and high precision, macro-micro composite drive schemes are often adopted, but they have the following prominent problems: First, the vacuum environment causes traditional lubrication to fail, and the "stick-slip" phenomenon and nonlinear hysteresis effect caused by dry friction are serious, making it difficult for conventional linear control algorithms to achieve stable and accurate nanoscale positioning; Second, the multi-degree-of-freedom manipulators introduced to achieve dexterous operation will produce significant "parasitic displacement" at the end effector due to the geometric errors and motion coupling of each joint, which is difficult to effectively compensate for through control algorithms in the SEM chamber where there is no external absolute measurement reference; Finally, the small space of the SEM chamber places extremely high demands on the system's integration, vacuum compatibility (low outgassing, non-magnetic, and resistance to high-temperature baking) and resistance to electron beam interference.

[0004] Therefore, there is an urgent need in this field for a new robot system that can adapt to the high vacuum environment of SEM, effectively solve the problems of nonlinear drive, motion coupling and cross-scale precision control, and achieve nanometer-level precise positioning and multi-degree-of-freedom dexterity operation over a large stroke range. Summary of the Invention

[0005] This invention provides a nano-vision micro-controlled robot system and its control method. The technical problem to be solved is: how to achieve stable and precise in-situ operation of micro-nano particles that are widely dispersed.

[0006] This invention provides a nano-vision micro-controlled robot system, comprising: a large-stroke motion module including a precision linear drive unit; a multi-degree-of-freedom manipulator disposed on the movable end of the precision linear drive unit, the multi-degree-of-freedom manipulator being used to adjust the attitude of the end effector, the end effector being used to perform operations on target particles; a micro-positioning module disposed on the large-stroke motion module and located within the operating range of the multi-degree-of-freedom manipulator, the micro-positioning module being used to place the target particles; and a control system electrically connected to the large-stroke motion module, the multi-degree-of-freedom manipulator, and the micro-positioning module respectively; the control system is configured to execute an adaptive control algorithm to achieve coordinated closed-loop control of the large-stroke motion module, the multi-degree-of-freedom manipulator, and the micro-positioning module, and to adjust the parameters of the adaptive control in real time through fuzzy logic to compensate for nonlinear errors.

[0007] Optionally, the micro-positioning module includes a Y-axis motion component and an X-axis motion component. The Y-axis motion component is fixedly connected to the large-stroke motion module, and the X-axis motion component is located on the movable end of the Y-axis motion component. The X-axis motion component is used to place the target particles.

[0008] Optionally, the multi-degree-of-freedom manipulator adopts a spherical coordinate system configuration, including: a second base rigidly connected to the large-stroke motion module; an azimuth rotation drive unit coaxially mounted on the second base, with its rotation axis set in the vertical direction to provide azimuth degree of freedom; a pitch rotation drive unit mounted on the azimuth rotation drive unit, with its rotation axis located in the horizontal plane and orthogonal to the rotation axis of the azimuth rotation drive unit to provide pitch degree of freedom; a radial telescopic drive unit rotatably mounted on the pitch rotation drive unit, with its telescopic axis perpendicular to the rotation axis of the pitch rotation drive unit to provide radial telescopic degree of freedom; and an end effector mounted on the radial telescopic drive unit.

[0009] Optionally, the control system includes a processor and a memory; the memory stores a kinematic control program, which, when executed by the processor, converts the Cartesian coordinates of the target operating point into joint control variables of each drive unit through inverse kinematics calculation.

[0010] Optionally, the long-stroke motion module includes: a first base, serving as the supporting body; a precision linear drive unit, mounted on the first base, which integrates a vacuum-compatible drive motor, a precision linear guide pair, and a drive controller; and a motion stage, slidably connected to the top of the precision linear drive unit, used to support the multi-degree-of-freedom manipulator and the micro-positioning module; wherein, the precision linear drive unit also has a high-precision grating ruler embedded inside, used to detect the actual physical position of the motion stage in real time and feed it back to the drive controller to construct a fully closed-loop position control circuit.

[0011] Optionally, both the X-axis motion assembly and the Y-axis motion assembly have a flexible hinge preload mechanism integrated inside.

[0012] Optionally, the control system is also configured to use an asymmetric waveform drive strategy to control the drive controller in the micro-positioning module; the asymmetric waveform is a segmented sawtooth wave control system that adjusts the asymmetry coefficient of the waveform in real time to match different load conditions.

[0013] This invention also provides a control method for a nanovision micro-controlled robot. The control method is executed by the nanovision micro-controlled robot system described above and includes the following steps: S100, Kinematic decoupling and spatial mapping: Obtain the Cartesian space target pose increment of the end effector of the multi-degree-of-freedom manipulator, and use inverse kinematic differential transformation to map the spatial error into the joint spatial error vector of each joint; S200, Hysteresis feedforward compensation: For the inherent asymmetric hysteresis nonlinearity of the piezoelectric ultrasonic guide, establish a hysteresis inverse model as a feedforward controller and output the feedforward control term; S300, Dynamic friction compensation: Introduce a dynamic friction model to describe the viscous-sliding friction phenomenon existing in the movement of the piezoelectric ultrasonic guide and calculate the friction compensation term; S400, Fuzzy adaptive feedback control: Using the joint spatial error vector as input, output the PID parameter correction amount in real time through fuzzy inference, and use an incremental PID algorithm to calculate the closed-loop feedback control amount; S500, Drive signal synthesis: Synthesize the feedforward compensation term, the friction compensation term and the feedback control amount to obtain the final drive control amount.

[0014] Optionally, in step S400, the fuzzy inference process further includes a micro-motion hold sub-step: when the absolute value of the system error is less than the nanometer-level positioning threshold, it automatically switches to the micro-motion hold mode; the micro-motion hold mode eliminates steady-state error by increasing the integral gain and simultaneously activates the dead-zone compensation voltage.

[0015] Optionally, in the micro-motion holding mode, the input variables include position error and error change rate, and the output variables include integral gain increment and dead zone compensation voltage.

[0016] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the control method of the nanoview micro-controlled robot as described above.

[0017] The present invention also provides a scanning electron microscope system, including a scanning electron microscope body and a nano-vision micro-control robot system as described above. The nano-vision micro-control robot system is disposed in the sample chamber of the scanning electron microscope body and is used to perform in-situ nano-manipulation in a vacuum environment.

[0018] This invention also provides a method for in-situ manipulation of micro / nano particles using the nano-vision micro-controlled robot system described above, comprising the following steps: mounting a sample stage carrying micro / nano particles onto a micro-positioning module; controlling a large-stroke motion module to move macroscopically, bringing the target particle into the operating range of a multi-degree-of-freedom manipulator; controlling the micro-positioning module to perform nanoscale precise positioning, placing the target particle within the predetermined operating area of ​​the end effector; controlling the multi-degree-of-freedom manipulator to adjust the spatial orientation of the end effector to the optimal operating angle; controlling the telescopic drive unit of the multi-degree-of-freedom manipulator to approach and contact the target particle in a micro-motion mode; thereby performing in-situ manipulation of the target particle; and coordinating the control of the large-stroke motion module, the multi-degree-of-freedom manipulator, and the micro-positioning module to transport the target particle to the target location.

[0019] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0020] This invention employs a composite configuration of a "large-stroke motion module, multi-degree-of-freedom manipulator, and micro-positioning module," driven by a collaborative control system integrating kinematic calculation, feedforward compensation, and fuzzy adaptive feedback, resulting in significant synergistic gains. Specifically, the large-stroke motion module provides macroscopic movement capabilities for target acquisition, the multi-degree-of-freedom manipulator offers flexible posture adjustment for approaching the target, and the micro-positioning module provides nanometer-level precise positioning for the final operation. Furthermore, the control system maps and distributes the Cartesian space error at the end effector to each joint through inverse kinematic calculation, effectively decomposing and compensating for parasitic displacements caused by joint coupling in the multi-degree-of-freedom manipulator. Simultaneously, by real-time tuning of PID parameters, it effectively addresses remaining unmodeled dynamics and disturbances, enabling the entire system to maintain extremely high control accuracy and stability even when facing multiple complex nonlinear factors such as vacuum dry friction, hysteresis, and motion coupling. Ultimately, the synergistic effect of all these technical features enables stable and precise in-situ manipulation of a wide range of micro / nano particles in harsh environments of high vacuum and confined spaces. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the structure of the nano-view micro-controlled robot involved in the present invention, as one embodiment.

[0022] Figure 2 This is a schematic diagram of the structure of the multi-degree-of-freedom manipulator of the Navview microcontroller in one embodiment.

[0023] Figure 3 This is a schematic diagram of the large-stroke motion module of the Navview micro-controlled robot in one embodiment.

[0024] Figure 4 This is a schematic diagram of the micro-positioning module of the Navview micro-control robot in one embodiment.

[0025] In the picture:

[0026] 1. Multi-degree-of-freedom manipulator; 11. Pitch angle rotation drive unit; 12. Azimuth angle rotation drive unit; 13. Second base; 14. Radial telescopic drive unit; 2. Large stroke motion module; 21. Motion stage; 22. First base; 23. Precision linear drive unit; 3. Microscopic positioning module; 31. Sample stage; 32. X-axis motion assembly; 33. Y-axis motion assembly. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.

[0028] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0029] In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims. In the description of this application, it should be understood that the terms "first," "second," "third," etc., are used only to distinguish similar objects and are not necessarily used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0030] Furthermore, in the description of this application, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0031] Example 1

[0032] In this embodiment, as Figure 1The illustrated nano-vision micro-controlled robot system includes: a long-stroke motion module 2, configured as a coarse positioning and transport platform; a multi-degree-of-freedom manipulator 1, mounted on the long-stroke motion module 2, used to adjust the attitude of the end effector, which is used to perform operations on the target particle; a micro-positioning module 3, mounted on the long-stroke motion module 2 and located within the operating range of the multi-degree-of-freedom manipulator 1, used to place the target particle; and a control system electrically connected to the long-stroke motion module 2, the multi-degree-of-freedom manipulator 1, and the micro-positioning module 3, respectively. The control system is configured to execute an adaptive control algorithm to achieve coordinated closed-loop control of the long-stroke motion module 2, the multi-degree-of-freedom manipulator 1, and the micro-positioning module 3, and to adjust the parameters of the adaptive control in real time through fuzzy logic to compensate for nonlinear errors.

[0033] In this embodiment, as Figure 2 As shown, the multi-degree-of-freedom manipulator 1 adopts an azimuth-pitch-radial spherical coordinate system configuration, specifically including:

[0034] The second base 13 is rigidly connected to the large stroke motion module 2;

[0035] The azimuth rotation drive unit 12 is coaxially mounted on the second base 13, and its rotation axis is set along the vertical direction (Z axis) to provide the azimuth degree of freedom θ1 in the horizontal plane;

[0036] Pitch angle rotation drive unit 11 is mounted on azimuth angle rotation drive unit 12. The rotation axis of pitch angle rotation drive unit 11 is located in the horizontal plane and is orthogonal to the rotation axis of azimuth angle rotation drive unit 12, and is used to provide pitch angle degree of freedom θ2.

[0037] The radial telescopic drive unit 14 is rotatably mounted on the pitch angle rotation drive unit 11. The telescopic axis of the radial telescopic drive unit 14 is perpendicular to the rotation axis of the pitch angle rotation drive unit 11, and is used to provide radial telescopic degree of freedom d3. The end effector is mounted on the radial telescopic drive unit 14.

[0038] In this embodiment, a forward kinematics model of the multi-degree-of-freedom manipulator 1 is established based on the Denavit-Hartenberg (DH) parameter method, and a homogeneous transformation matrix is ​​defined from the coordinate system {0} of the first base 22 to the coordinate system {3} of the end effector. Its calculation formula is:

[0039] Formula (1);

[0040] in:

[0041] The homogeneous transformation matrix from coordinate system {0} to coordinate system {3} is the standard representation in the DH parameter method;

[0042] In this context, 0 and 3 are both subscripts, representing the base coordinate system {0} (i.e., the first base 22 coordinate system) and the end coordinate system {3} (i.e., the end effector coordinate system), respectively.

[0043] The transformation matrix representing joint 1 (azimuth rotation);

[0044] The transformation matrix representing joint 2 (pitch angle rotation);

[0045] This represents the transformation matrix of joint 3 (radial stretching).

[0046] The spatial position coordinates (Px, Py, Pz) of the end effector in the coordinate system of the first base 22 are determined by the following spherical coordinate kinematic equations:

[0047] Formula (2);

[0048] Formula (3);

[0049] Formula (4);

[0050] in:

[0051] The rotation angle of the azimuth rotation drive unit 12 relative to the X-axis;

[0052] The elevation angle of the pitch angle rotation drive unit 11 relative to the horizontal plane;

[0053] The real-time extension length of the radial telescopic drive unit 14;

[0054] The initial arm length constant is: when the radial telescopic drive unit 14 is at zero position. At that time, the horizontal distance from the center of the pitch rotation axis to the tip of the end effector needle;

[0055] The vertical height constant is the vertical height of the pitch rotation axis center relative to the bottom surface of the second base 13.

[0056] In this embodiment, the control system includes a processor and a memory; the memory stores a kinematic control program, and when the kinematic control program is executed by the processor, the Cartesian coordinates of the target operation point are converted into joint control variables of each drive unit through inverse kinematics calculation.

[0057] Specifically, the kinematic control program converts the Cartesian coordinates (x, y, z) of the target manipulator point into joint control variables for each drive unit through inverse kinematics calculation. This allows for precise spatial positioning of the sample from all angles.

[0058] In this embodiment, as Figure 3 As shown, the long-stroke motion module 2 includes: a first base 22, serving as the supporting body; a precision linear drive unit 23, which integrates a vacuum-compatible drive motor, a guide rail body, a precision linear guide pair, and a drive controller; and a motion stage 21, slidably connected to the top of the precision linear drive unit 23, which is used to support the multi-degree-of-freedom manipulator 1. The precision linear drive unit 23 also has a high-precision grating ruler embedded inside, which is used to detect the actual physical position of the motion stage 21 in real time and feed it back to the drive controller to construct a fully closed-loop position control circuit.

[0059] It should be noted that the precision linear drive unit 23, which integrates a vacuum-compatible drive motor, guide rail body, precision linear guide pair and drive controller, is a commercially available product, and its structural principle will not be described in detail here.

[0060] In this embodiment, the first base 22 serves as the supporting body and is made of high-rigidity material, with an installation interface at its bottom. A precision linear drive unit 23 is mounted on the first base 22, and integrates a vacuum-compatible drive motor, a guide rail body, a precision linear guide pair, and a corresponding drive controller. The reading head of the grating ruler moves synchronously with the motion stage 21, and the scale is fixed to the guide rail body for real-time detection of the actual physical position of the motion stage 21.

[0061] In this embodiment, the control system constructs a fully closed-loop position control loop based on the feedback signal from the grating ruler. The specific control logic is as follows:

[0062] The control system receives the target position command and drives the drive motor in the precision linear drive unit 23 to operate.

[0063] The grating ruler collects the displacement data of the motion table 21 in real time and feeds it back to the drive controller;

[0064] The drive controller compares the actual position fed back by the grating ruler with the target position, calculates the position deviation, and corrects the output torque and speed of the drive motor in real time, thereby eliminating transmission errors caused by thermal deformation of the lead screw or mechanical backlash.

[0065] In this embodiment, the motor windings, grating ruler reading head and drive circuit board in the precision linear drive unit 23 are all subjected to vacuum low-outgas treatment, and the guide pair is lubricated with vacuum-specific solid lubricant or low vapor pressure grease to adapt to the high vacuum working environment.

[0066] In this embodiment, the control system is also configured to use an asymmetric waveform driving strategy to control the drive controller in the micro-positioning module 3; the asymmetric waveform is a segmented sawtooth wave. The control system adjusts the asymmetry coefficient of the waveform in real time to match different load conditions.

[0067] In this embodiment, as Figure 4 As shown, the micro-positioning module 3 includes a Y-axis motion component 33 and an X-axis motion component 32. The Y-axis motion component 33 is fixedly connected to the large stroke motion module 2, and the X-axis motion component 32 is disposed on the movable end of the Y-axis motion component 33. The X-axis motion component 32 is used to place the target particles.

[0068] In this embodiment, both the X-axis motion component 32 and the Y-axis motion component 33 are ultrasonic guide rail units, and both the X-axis motion component 32 and the Y-axis motion component 33 integrate a flexible hinge pre-tensioning mechanism. The flexible hinge pre-tensioning mechanism can effectively reduce or eliminate backlash, thereby improving the resolution and response speed of micro-positioning.

[0069] In this embodiment, a flexible hinge preload mechanism is used to replace the traditional rigid spring preload; the control system performs topology optimization on the flexible hinge preload mechanism based on Castigliano's second theorem to obtain extremely low motion direction stiffness. and extremely high parasitic directional stiffness Its stiffness ratio optimization objective function Defined as:

[0070] Formula (5);

[0071] Among them, the stiffness in the direction of motion The theoretical derivation formula is as follows:

[0072] Formula (6);

[0073] in:

[0074] E is the elastic modulus of the material;

[0075] b is the hinge width;

[0076] t is the thickness at the thinnest point;

[0077] L is the hinge length;

[0078] r is the radius of the fillet;

[0079] This is a dimensionless compliance correction factor.

[0080] In this embodiment, by making Minimize to reduce or eliminate mechanical backlash, thereby achieving a highly sensitive response to nanometer-scale changes in drive voltage.

[0081] In this embodiment, the drive controller is a piezoelectric ceramic controller.

[0082] In this embodiment, the control system uses an asymmetric sawtooth wave as the driving source for the piezoelectric ceramic controller. By optimizing the duty cycle and falling edge slope of the waveform, the load capacity of the stick-slip drive is improved.

[0083] Specifically, by establishing a driving force transmission efficiency model, the effective net driving force within a single cycle is defined. for:

[0084] Formula (7);

[0085] in piezoelectric driving force Friction

[0086] To maximize To drive a large load, the control system optimizes the drive voltage waveform U(t) into a piecewise function, as follows:

[0087] (1) Stick Phase:

[0088] Formula (8);

[0089] At this point, the piezoelectric ceramic slowly elongates, utilizing static friction. Drives the load to move;

[0090] (2) Slip Phase:

[0091] Formula (9);

[0092] At this moment, the piezoelectric ceramic rapidly retracts, generating inertial force. The force is greater than the maximum static friction force, enabling slippage and recovery.

[0093] in This is the maximum value of the driving voltage. To allow for a gradual increase over a period of time, This is a period of rapid descent;

[0094] The control system adjusts the waveform asymmetry coefficient in real time. The calculation formula is as follows:

[0095] Formula (10).

[0096] In this embodiment, the waveform asymmetry coefficient is adjusted. The specific strategy is as follows:

[0097] According to load quality Friction properties ,

[0098] Formula (11);

[0099] piezoelectric ceramic transient acceleration Depend on and Decide.

[0100] In this embodiment, by setting To ensure the inertial force generated during the rapid descent, the load should be appropriately increased when the load increases. (Extend the gradual ascent phase and shorten the rapid descent phase) to improve At the same time, it avoids mechanical resonance that leads to vibration instability, thereby achieving high load driving capability at nanometer-level resolution.

[0101] Example 2

[0102] This embodiment provides a control method for a nano-vision micro-control robot, based on Embodiment 1.

[0103] In this embodiment, the control method is executed by the nanovision micro-controlled robot system as described above, and includes the following steps:

[0104] S100, Kinematic Decoupling and Spatial Mapping: Obtain the Cartesian space target pose increment of the end effector of the multi-degree-of-freedom manipulator 1, and use inverse kinematic differential transformation to map the spatial error into the joint spatial error vector of each joint;

[0105] S200 Hysteresis Feedforward Compensation: To address the inherent asymmetric hysteresis nonlinearity of piezoelectric ultrasonic guides, a hysteresis inverse model is established as a feedforward controller, and a feedforward control term is output.

[0106] S300, Dynamic Friction Compensation: A dynamic friction model is introduced to describe the viscous-sliding friction phenomenon that exists during the movement of the piezoelectric ultrasonic guide rail, and the friction compensation term is calculated.

[0107] S400, Fuzzy Adaptive Feedback Control: Using the joint space error vector as input, it outputs the PID parameter correction amount in real time through fuzzy inference and uses the incremental PID algorithm to calculate the closed-loop feedback control amount;

[0108] S500, Drive Signal Synthesis: Combines the feedforward compensation term, friction compensation term, and feedback control quantity to obtain the final drive control quantity.

[0109] Specifically, by integrating feedforward compensation based on the hysteresis inverse model and dynamic friction compensation based on the dynamic friction model, the main nonlinear factors of piezoelectric drive are pre-emptively offset. Furthermore, by combining fuzzy adaptive feedback control, the PID parameters are tuned in real time to cope with the remaining unmodeled dynamics and residual disturbances. This composite control strategy of feedforward compensation for the main nonlinearity and feedback suppression of residual disturbances enables the entire system to still exhibit extremely high control accuracy and stability when facing multiple complex nonlinear factors such as vacuum dry friction, hysteresis, and motion coupling.

[0110] In this embodiment, the target particles are micro- and nano-particles.

[0111] In this embodiment, in step S100, kinematic decoupling and spatial mapping are based on the RRP configuration.

[0112] Specifically, for the dual-rotation and single-telescopic cascade configuration of the multi-degree-of-freedom manipulator 1, consisting of the azimuth rotation drive unit 12, the pitch rotation drive unit 11, and the radial telescopic drive unit 14, a forward kinematic model is established; the control system receives the Cartesian space target pose increment from the end effector. The spatial error is mapped to the joint spatial error vector of each joint using inverse kinematic differential transformation. The calculation formula is:

[0113] Formula (12);

[0114] in:

[0115] q represents the current joint variable direction;

[0116] J -1 (q) is the Jacobian inverse matrix.

[0117] When the determinant of the Jacobian matrix det(J) approaches zero, the inverse matrix is ​​solved by damped least squares method to avoid singular configurations, thereby achieving independent decoupling control of multi-degree-of-freedom motion.

[0118] In this embodiment, in step S200, considering the inherent asymmetric hysteresis nonlinearity of the piezoelectric ultrasonic guide, a hysteresis inverse model based on the Prandtl-Ishlinskii (PI) operator is established as a feedforward controller. This allows for the pre-linearization of the input-output relationship of the piezoelectric ceramic during the open-loop phase.

[0119] Specifically, the feedforward controller is based on the target displacement r （k） Calculate the feedforward voltage U used to counteract the hysteresis effect. ff (k), whose discretized expression is:

[0120] Formula (13);

[0121] in:

[0122] The linear gain coefficient;

[0123] These are the weighting coefficients;

[0124] For the threshold;

[0125] This represents the summation over all hysteresis operator units.

[0126] In this embodiment, the parameters of the PI model (hysteresis inverse model) are determined through the following experimental identification process: a quasi-static triangular wave voltage with increasing amplitude is applied to the piezoelectric ceramic, and the input voltage and output displacement data are collected synchronously using a laser displacement sensor. 1000 data points are collected and the average value is taken after repeating 5 times.

[0127] Linear gain coefficient The slope of the main diagonal is obtained by fitting the slope using the least squares method; the main diagonal (Major Diagonal / Principal Diagonal) refers to the ascending curve. With the descending curve The average value, which represents the "centerline" or "average response" of the hysteresis loop, is calculated using the following formula:

[0128] Formula (14);

[0129] The main diagonal is the axis of symmetry of the hysteresis loop (in the case of ideal symmetrical hysteresis). It eliminates the influence of the hysteresis effect and reflects the average linear characteristics of the piezoelectric ceramic. The slope of the main diagonal is the linear gain coefficient. .

[0130] The slope of the main diagonal is calculated using the direct averaging method, which includes the following steps:

[0131] (1) Collect voltage rise curve data: k=1,2,...,N;

[0132] (2) Collect voltage drop curve data: {( )}, k=1,2,...,N;

[0133] (3) Calculate the main diagonal data:

[0134] Formula (15);

[0135] (4) Fit the linear relationship using the least squares method:

[0136] Formula (16).

[0137] threshold r i The calculation method is as follows: the displacement range is divided into M equal intervals, and the threshold sequence is:

[0138] Formula (17).

[0139] Weighting coefficients { By solving the least squares optimization problem, we obtain:

[0140] Formula (18).

[0141] in:

[0142] This represents the k-th voltage value during the voltage rise phase.

[0143] This represents the k-th displacement value during the voltage rise segment;

[0144] This represents the k-th voltage value in the voltage drop segment.

[0145] This represents the k-th displacement value during the voltage drop segment.

[0146] i is the index of the hysteresis operator (the i-th threshold point);

[0147] This represents the maximum displacement during the experiment;

[0148] y(k) is the actual output displacement of the kth sampling point;

[0149] r(k) represents the target displacement / command displacement of the kth sampling point.

[0150] In this embodiment, in step S300, to address the viscous-sliding friction phenomenon during the guide rail movement, the LuGre dynamic friction model is introduced to describe the frictional force, and the friction compensation voltage U is calculated accordingly. fric .

[0151] Specifically, the LuGre model parameters are identified and determined through friction characteristic experiments based on the specific guide rail structure, material contact state, and operating conditions, and can be dynamically corrected during system operation to improve the accuracy and stability of friction compensation.

[0152] Model defines internal friction state variables The differential equation is:

[0153] Formula (19);

[0154] The final calculated friction compensation term is:

[0155] Formula (20);

[0156] in:

[0157] It is the relative velocity;

[0158] This is the stiffness coefficient;

[0159] This is the Stribeck effect function.

[0160] The damping coefficient;

[0161] It is the coefficient of viscous friction.

[0162] In this embodiment, in step S400, the fuzzy adaptive feedback control employs an incremental PID (Proportional-Integral-Derivative) controller. The incremental PID uses the joint space error vector E obtained in step S100. q Taking each component of (k) as input, the components of the joint space error vector are represented as follows:

[0163] Formula (21);

[0164] in:

[0165] e j (k) represents the position error of the j-th joint in the k-th control cycle, where k is the discrete time index and T is the sampling period.

[0166] e j The rate of change of error (k) is defined as:

[0167] Formula (22);

[0168] The PID parameter correction is output in real time through fuzzy inference. The PID parameter correction includes the proportional gain increment (ΔK). p ), Integral gain increment (ΔK) i ) and differential gain increment (ΔK) d The incremental PID algorithm is used to calculate the closed-loop feedback control quantity. An independent PID controller is designed for each joint, and the control law of the j-th joint is... for:

[0169] Formula (23).

[0170] In this embodiment, in step S400, the fuzzy inference process further includes a micro-motion holding sub-step: when the absolute value of the system error of the nanoview micro-control robot is less than the nanometer-level positioning threshold, it automatically switches to the micro-motion holding mode; the micro-motion holding mode eliminates steady-state error by increasing the integral gain and simultaneously activates the dead zone compensation voltage.

[0171] Specifically, when the system error satisfies When the time is reached, it automatically switches to micro-motion hold mode. In micro-motion hold mode, the integral gain increment ΔK is increased. i Eliminate steady-state error and simultaneously activate dead-zone compensation voltage U. dead The input variables include the position error e(k) (representing the position error of a single joint, which is a scalar) and the rate of change of error Δe(k); the output variables include the integral gain increment ΔK. i and dead zone compensation voltage U dead The input and output variables are divided into fuzzy sets (such as NB, NS, Z, PS, and PB representing the five fuzzy levels of error e(k) and error change rate Δe(k): negative large, negative small, zero, positive small, and positive large, while Low, Medium, and High correspond to the three fuzzy states of the proportional coefficient and dead zone parameter, respectively. This fuzzification of linguistic variables lays the foundation for the formulation of subsequent fuzzy control rules).

[0172] Specifically, the core rules of fuzzy sets are exemplified as follows:

[0173] When e(k)≈Z and Δe(k)≈Z, ΔK i For High, U deadThe value is Medium, and the remaining intervals are adjusted accordingly based on the error magnitude and rate of change. The fuzzy inference adopts the Mamdani type, and the defuzzification adopts the centroid method. The control performance can be optimized by online or offline fine-tuning based on the steady-state error and jitter amplitude of the experimental micro-motion positioning.

[0174] In this embodiment, in step S500, the drive signal is obtained by combining the feedforward compensation term, the friction compensation term, and the feedback control term.

[0175] Specifically, the drive signal is the drive control quantity U. total (k), its calculation formula is as follows:

[0176] Formula (24);

[0177] In this embodiment, the control system according to The magnitude of the amplitude determines the execution of the dual-mode driving strategy:

[0178] (1) Macro mode: When the control quantity is large, a pulse signal (PFM) with fixed amplitude and variable frequency is output, and the resonance characteristics of the ultrasonic motor are used to achieve rapid large stroke movement;

[0179] (2) Micro-motion mode: When the control quantity is small, the drive frequency is locked, and the drive voltage amplitude (AM) or pulse duty cycle (PWM) is adjusted. The inverse piezoelectric effect of piezoelectric ceramics is used to achieve sub-nanometer quasi-static deformation, thereby completing precise positioning at the near-atomic scale.

[0180] Example 3

[0181] This embodiment provides a computer-readable storage medium based on any of the above embodiments.

[0182] In this embodiment, a computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, it implements the control method of the nanoview micro-controlled robot as described above.

[0183] Example 4

[0184] This embodiment provides a scanning electron microscope system based on any of the above embodiments.

[0185] In this embodiment, the scanning electron microscope system includes a scanning electron microscope body and a nano-vision micro-control robot system as described above. The nano-vision micro-control robot system is disposed in the sample chamber of the scanning electron microscope body and is used to perform in-situ nano-manipulation in a vacuum environment.

[0186] Example 5

[0187] This embodiment provides a method for in-situ manipulation of micro-nano particles using the aforementioned nano-vision micro-control robot system.

[0188] In this embodiment, the in-situ manipulation of micro / nano particles includes the following steps:

[0189] The sample stage 31 carrying micro-nano particles is mounted on the micro positioning module 3.

[0190] The large-stroke motion module 2 is controlled to move macroscopically, so that the target particle enters the operating range of the multi-degree-of-freedom manipulator 1;

[0191] The micro-positioning module 3 is controlled to perform nanometer-level precise positioning, so that the target particle is in the predetermined operating area of ​​the end effector;

[0192] Control the multi-degree-of-freedom manipulator 1 to adjust the spatial orientation of the end effector to the optimal operating angle;

[0193] The telescopic drive unit of the multi-degree-of-freedom manipulator 1 is controlled to approach and contact the target particle in a micro-motion mode;

[0194] This allows for in-situ manipulation of the target particles;

[0195] The coordinated control of the large-stroke motion module 2, the multi-degree-of-freedom manipulator 1, and the micro-positioning module 3 transports the target particles to the target position.

[0196] The mechanical property testing method for micro / nano materials in this embodiment adopts the end-to-end intelligent manipulation method of SEM as described above, which controls the micro / nano manipulation probe to perform compression, tension or bending operations on the target micro / nano material, while using a scanning electron microscope to observe the deformation response of the material.

[0197] Specifically, the nanovision micro-controlled robot system in this embodiment is applied to a scanning electron microscope to conduct in-situ manipulation experiments of metal-organic framework (MOF) micro-nano particles inside the SEM chamber.

[0198] In this embodiment, the azimuth rotation drive unit 12 and the pitch rotation drive unit 11 employ vacuum-compatible servo motors. Their windings use high-temperature resistant vacuum enameled wire, and the bearings use vacuum grease or solid lubrication. The housing structure is designed with a low outgassing rate material and undergoes vacuum baking degassing treatment. Simultaneously, the encoders of the azimuth rotation drive unit 12 and the pitch rotation drive unit 11 employ an anti-electromagnetic interference design; if necessary, a metal shield can be added to prevent interference from the SEM electron beam.

[0199] In this embodiment, MOF micro / nano particles with a particle size of 500 nm to 2 μm were selected as the manipulated object, and the target particles were uniformly dispersed on the surface of a silicon wafer carrier with dimensions of 10 mm × 10 mm × 0.2 mm. To improve SEM imaging quality and sample conductivity, the carrier was coated with a 5 nm gold film and fixed to the stage of the micro-positioning module 3 using conductive adhesive to keep it stable during in-situ operation.

[0200] In this embodiment, the end effector uses a tungsten needle with a tip radius of approximately 80-120 nm.

[0201] Specifically, driven by the multi-degree-of-freedom manipulator 1 and the large-stroke motion module 2, and in conjunction with the movement of the micro-positioning module 3, the tungsten needle pushes the target particle to the target position in the SEM chamber.

[0202] In other embodiments, the end effector may also be configured as a flexible clamp or a micro-cutting tool, depending on the needs of different tasks.

[0203] In this embodiment, after the entire nanovision micro-controlled robot system is placed into the SEM chamber, the connection between the internal drive circuitry and the external controller is established via an air-mounted flange. The scanning electron microscope completes the vacuuming (10... -3 ~10 -5 After P is reached, the nano-view micro-controlled robot can operate in in-situ micromanipulation mode. First, a global scan is performed using a large-stroke platform to bring the silicon wafer carrier into the SEM field of view; then, the X-axis motion component 32 and Y-axis motion component 33 of the micro-positioning module 3 are used for nanoscale position correction to place the target particle within the predetermined operating area of ​​the needle tip. To ensure the needle tip has the optimal approach angle, the drive controller adjusts the two-stage rotary joints θ1 and θ2 of the robot arm so that the needle tip axis forms an incident angle of approximately 30°~45° with the sample plane. The positioning process of the needle tip is completed through inverse kinematics solution. Specifically, given the target position P of the needle tip... target =[x t , y t , z t ] T The control system calculates the required joint angles using the following analytical solution:

[0204] (1) Azimuth angle calculation:

[0205] Formula (25);

[0206] (2) Calculation of radial expansion / contraction:

[0207] First, calculate the horizontal projection distance. Total arm length :

[0208] Formula (26);

[0209] Formula (27);

[0210] The radial expansion / contraction is then calculated as follows: Formula (28).

[0211] (3) Solving for the pitch angle:

[0212] Formula (29).

[0213] After the solution is obtained, the values ​​of azimuth, radial extension and pitch angle are substituted into the forward kinematic model of formula (1) to verify the accuracy of the solution. The specific verification method is as follows:

[0214] Extracting the position components of the transformation matrix [P] x , P y , P z ] T Calculate the deviation from the target position. :

[0215] Formula (31);

[0216] in:

[0217] The actual value at the target location;

[0218] The calculated value for the target location;

[0219] when When the value is less than 50 nm, the inverse kinematics solution is considered valid, driving each joint to move to the target angle. The control system compensates for possible parasitic displacements and thermal drifts in real time using the Jacobian matrix, keeping the actual position drift of the tungsten needle tip within 50 nm.

[0220] As the needle tip approaches the micro / nano particles, the drive controller activates a micro-motion mode to achieve final contact between the needle tip and the micro / nano particles at an extremely low speed of approximately 20 nm / s.

[0221] This embodiment verifies the stability of the platform of the present invention in performing high-precision in-situ operation and repositioning of micro and nano particles under high vacuum conditions, and further calibrates the contact force model through the correspondence between the theoretical model and experimental measurements.

[0222] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A nano-vision micro-controlled robot system, characterized in that, The application relates to a multi-freedom-degree manipulator, which comprises the following parts: a long-stroke motion module (2) comprising a precision linear driving unit (23); a multi-freedom-degree manipulator (1) arranged on the moving end of the precision linear driving unit (23), wherein an end effector is arranged on the multi-freedom-degree manipulator (1), the multi-freedom-degree manipulator (1) is used for adjusting the posture of the end effector, and the end effector is used for performing an operation on a target particle; a micro-positioning module (3) arranged on the long-stroke motion module (2) and located in the operation range of the multi-freedom-degree manipulator (1), wherein the micro-positioning module (3) is used for placing the target particle; a control system electrically connected with the long-stroke motion module (2), the multi-freedom-degree manipulator (1) and the micro-positioning module (3) respectively, wherein the control system is configured to execute an adaptive control algorithm to realize the collaborative closed-loop control of the long-stroke motion module (2), the multi-freedom-degree manipulator (1) and the micro-positioning module (3), and compensate for nonlinear errors by adjusting the parameters of the adaptive control in real time through fuzzy logic; the control system comprises a processor and a memory, and a kinematic control program is stored in the memory, wherein when the kinematic control program is executed by the processor, the Cartesian coordinates of a target operation point are converted into joint control variables of the multi-freedom-degree manipulator (1) through inverse kinematic calculation; the control system is further configured to control a driving controller in the micro-positioning module (3) by adopting an asymmetric waveform driving strategy; the asymmetric waveform is a segmented sawtooth wave, and the control system adjusts the asymmetry coefficient of the waveform in real time to match different load conditions; the control system optimizes a driving voltage waveform U (t) into a segmented function, and the segmented function is as follows: (1) a stick phase: ； At this time the piezoelectric ceramic slowly extends, and the static friction force is utilized To drive the load to move; (2) a slip phase: ； At this time, the piezoelectric ceramic quickly retracts, and the inertial force generated Is greater than the maximum static friction force, realizing slip reset; wherein is the maximum driving voltage, is the slow ramp time, is the fast ramp time; A control system adjusts a waveform asymmetry coefficient in real time The calculation formula is as follows: 。 2. The nanoscopic robotic system of claim 1, wherein, the micro-positioning module (3) comprises a Y-axis motion assembly (33) and an X-axis motion assembly (32), wherein the Y-axis motion assembly (33) is fixedly connected to the long-stroke motion module (2), and the X-axis motion assembly (32) is arranged on the moving end of the Y-axis motion assembly (33) and is used for placing the target particle.

3. The nanoscopic robotic system of claim 1, wherein, the multi-freedom-degree manipulator (1) adopts a spherical coordinate system configuration, and comprises: a second base (13) rigidly connected to the precision linear driving unit (23); an azimuth angle rotary driving unit (12) coaxially installed on the second base (13), wherein the rotary axis of the azimuth angle rotary driving unit (12) is arranged in a vertical direction and is used for providing an azimuth angle freedom degree; a pitch angle rotary driving unit (11) installed on the azimuth angle rotary driving unit (12), wherein the rotary axis of the pitch angle rotary driving unit (11) is located in a horizontal plane and is orthogonal to the rotary axis of the azimuth angle rotary driving unit (12) and is used for providing a pitch angle freedom degree. A radial telescopic drive unit (14) is rotatably mounted on a pitch angle rotation drive unit (11). The telescopic axis of the radial telescopic drive unit (14) is perpendicular to the rotation axis of the pitch angle rotation drive unit (11) to provide radial telescopic freedom. The end effector is mounted on the radial telescopic drive unit (14).

4. The nanoscopic robotic system of claim 1, wherein, The long-stroke motion module (2) includes: The first base (22) serves as the main supporting structure; A precision linear drive unit (23) is mounted on the first base (22) and integrates a vacuum-compatible drive motor, a precision linear guide pair and a drive controller. The motion table (21) is slidably connected to the top of the precision linear drive unit (23) and is used to support the multi-degree-of-freedom manipulator (1). The precision linear drive unit (23) also has an embedded grating ruler inside. The grating ruler is used to detect the actual physical position of the motion table (21) in real time and feed it back to the drive controller to build a fully closed-loop position control loop.

5. The nanoscopic robotic system of claim 2, wherein, Both the X-axis motion component (32) and the Y-axis motion component (33) have a flexible hinge pre-tensioning mechanism integrated inside.

6. A control method of a nano vision micro controlled robot, characterized in that, The control method is executed by the nanovision micro-control robot system as described in any one of claims 1 to 5, and includes the following steps: S100, Kinematic decoupling and spatial mapping: Obtain the Cartesian space target pose increment of the end effector of the multi-degree-of-freedom manipulator (1), and use inverse kinematic differential transformation to map the spatial error into the joint spatial error vector of each joint; S200 Hysteresis Feedforward Compensation: To address the inherent asymmetric hysteresis nonlinearity of piezoelectric ultrasonic guides, a hysteresis inverse model is established as a feedforward controller, and a feedforward control term is output. S300, Dynamic Friction Compensation: A dynamic friction model is introduced to describe the viscous-sliding friction phenomenon that exists during the movement of the piezoelectric ultrasonic guide rail, and the friction compensation term is calculated. S400, Fuzzy Adaptive Feedback Control: Using the joint space error vector as input, the PID parameter correction is output in real time through fuzzy inference, and the closed-loop feedback control quantity is calculated using the incremental PID algorithm; S500, Drive Signal Synthesis: Combines the feedforward compensation term, friction compensation term, and feedback control quantity to obtain the final drive control quantity.

7. The control method according to claim 6, characterized by, In step S400, the fuzzy inference process further includes a micro-motion hold sub-step: when the absolute value of the system error is less than the nanometer-level positioning threshold, it automatically switches to the micro-motion hold mode; the micro-motion hold mode eliminates steady-state error by increasing the integral gain and simultaneously activates the dead-zone compensation voltage.

8. The control method according to claim 7, characterized by, In the micro-motion holding mode, the input variables include position error and error change rate, and the output variables include integral gain increment and dead zone compensation voltage.

Images