Ultrasonic robot tail end zero-rigidity constant-force driver and control method thereof

Abstract

The invention discloses an ultrasonic robot tail end zero-rigidity constant-force driver and a control method thereof.The ultrasonic robot tail end zero-rigidity constant-force driver comprises a shell, a negative-rigidity structure, a positive-rigidity structure and a control circuit board, the shell comprises an external main frame, a top cover and a base, the negative-rigidity structure comprises a fixed magnet and a movable magnet set, and the positive-rigidity structure comprises a positive-rigidity structure and a negative-rigidity structure; the fixed magnet is fixedly installed on the base, the movable magnet set is fixedly installed on the top cover, the positive stiffness structure comprises a linear servo motor, a displacement sensor, a linear spring and a spring seat, the movable magnet set and the fixed magnet generate negative stiffness through magnetic field interaction, and the linear spring generates positive stiffness through deformation. And the resultant forces are offset to realize zero-rigidity constant-force output. The control circuit board is installed in the base and electrically connected with the linear servo motor and the displacement sensor. The method is used for solving the problems that in the prior art, scanning efficiency is low, the method is sensitive to measurement errors, and ultrasonic scanning requirements of complex scenes are difficult to meet.

Description

Technical Field

[0001] The present invention relates to the technical field of medical robots, and in particular to a zero-stiffness constant-force driver at the end of an ultrasonic robot and a control method thereof. Background Art

[0002] Since the contact between the ultrasound probe and the human body is a prerequisite for ensuring the effective propagation of ultrasound signals and thus achieving high-quality ultrasound imaging, and the patient's body surface contours are irregular and vary significantly from patient to patient, the ultrasound scanning robot cannot perform ultrasound scans on different patients according to a fixed scanning path, and reasonable compensation for breathing must be considered. The existing technology requires time-consuming and multiple refined ultrasound scanning path planning based on the patient's contour characteristics before performing the ultrasound scan. This not only has low fault tolerance, but also greatly reduces ultrasound scanning efficiency. Therefore, the ultrasound robot needs to have more flexible path planning capabilities.

[0003] During ultrasound scanning, excessive or insufficient scanning force between the ultrasound probe and the patient will lead to a decrease in ultrasound imaging quality. The scanning area close to the bones has a large stiffness gradient, and a slight motion deviation will cause drastic changes in the scanning force between the robot and the patient, posing a threat to the patient's safety. Therefore, actuators and corresponding force-position control strategies suitable for human ultrasound scanning must be designed for clinical scenarios to achieve stable scanning force between the ultrasound probe and the patient. However, clinical scenarios are complex and changeable, and existing rigid actuators and control strategies are difficult to meet the complex clinical needs, and safety is difficult to ensure. Therefore, ultrasound robots need to have more stable force-position control capabilities.

[0004] Achieving fast and accurate trajectory tracking control with time constraints at the end of the robotic arm is also a problem to be solved. This is because the robotic arm system is a highly nonlinear and strongly coupled dynamic system, and in practical applications there are inevitable uncertainties such as measurement errors, external disturbances, and dynamic modeling errors, which will have a significant impact on the trajectory tracking control of the robotic arm. Existing trajectory tracking control algorithms generally have the problem of actuator saturation system jitter, which poses a potential threat to the safety of equipment and patients. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a zero-stiffness constant-force driver at the end of an ultrasonic robot and a control method thereof, which are used to solve the problems raised in the background technology that the existing technology has low scanning efficiency, measurement errors, and is difficult to adapt to the ultrasonic scanning needs of complex scenes.

[0007] (2) Technical solution

[0008] To achieve the above object, the present invention provides the following technical solution: An ultrasonic robot end zero-stiffness constant-force driver, comprising a housing, a positive stiffness structure, a negative stiffness structure, and a control circuit board.

[0009] The housing includes an external main frame, a top cover, a base, and internal bearings. The base and the top cover are connected by an optical axis and linear bearings.

[0010] The negative stiffness structure includes a fixed magnet, a moving magnet group, and a fixed magnet holder. The fixed magnet is fixed to the base through the fixed magnet holder. The moving magnet group includes magnets and aluminum blocks, which are alternately stacked. The magnets and the aluminum blocks are connected to the grooves of the top cover by screws.

[0011] The top cover is connected to the optical axis through the optical axis fixing holes. Each optical axis contains two linear bearings. The linear bearings are fixed to the bearing seats through the bearing holes, and the bearing seats are fixed to the base through the screw holes.

[0012] The linear servo motor is connected to the base by screws. The motor rod is connected to the spring seat by screws. The spring is fixed by the top cover and the spring seat.

[0013] The displacement sensor is connected to the base through the sensor snap ring, and the sensor rod is connected to the top cover.

[0014] On the other hand, the present application also provides an ultrasonic robot, comprising an ultrasonic robot end zero-stiffness constant-force driver in the above technical solution, further comprising a six-axis robotic arm, an ultrasonic probe, a six-axis force sensor, a 3D camera, and a host computer system. The ultrasonic robot end zero-stiffness constant-force driver is installed at the end of the six-axis robotic arm; the ultrasonic probe, the six-axis force sensor, and the 3D camera are integrated on the ultrasonic robot end zero-stiffness constant-force driver; the host computer system is used to realize data acquisition, trajectory planning, and real-time operation of the adaptive control algorithm. The driver adjusts the restoring force curve through a feedforward control mechanism in combination with an electromagnetic actuator to achieve large-stroke constant-force output. The host computer system is developed based on Qt and includes a communication module, a controller module, a point cloud trajectory planning module, and an inverse kinematics module.

[0015] A control method for an ultrasonic robot end zero-stiffness constant-force driver in the technical solution is jointly composed of the control circuit board, the linear servo motor, and the displacement sensor. The control part of the entire system is mainly divided into a position control inner loop and an impedance control outer loop. The inner loop control uses the PID method.

[0016] In the technical solution, for the external input environment, the contact force between the ultrasonic probe at the end of the robot and the medical dummy is obtained by a six-axis force sensor, and the force deviation is obtained by subtracting the expected force. Combining the second derivative of the displacement sensor data, the contact force deviation caused by the inertial force of the ultrasonic probe at the end is obtained. Then, the sum of the two force deviations is input into the impedance controller to generate a position correction amount, which corrects the reference trajectory points of the trajectory planning, so as to achieve the purpose of compliant control.

[0017] The trajectory planning is realized through the following steps: using a 3D camera to collect target point cloud data, and extracting path points containing normal vectors after preprocessing; extracting path points within a range of 1 mm width at intervals of 50 mm, and using the spline interpolation method to generate 5000 smooth trajectory points; combining the bilinear discretization method to convert the adaptive control algorithm from the Laplace domain to the discrete domain, and calculating the position deviation in real time.

[0018] In the technical solution, the selection of the impedance control parameters needs to consider the steady-state effect and the dynamic response of the system. Considering the feedforward-feedback hybrid control of flexible drive, the system has the zero stiffness characteristic, so the transfer function has the simplest form, which simplifies the control algorithm and is beneficial to the improvement of robustness.

[0019] In order to prevent the sudden change of the axial force in case of emergencies, while using the zero stiffness characteristic for passive adjustment in the constant force interval, a proportional-derivative control of the contact force is introduced into the impedance model for active adjustment and to reduce the influence of the sudden change of inertial force.

[0020] Under the influence of the feedforward-feedback hybrid control, the displacement influence of the system chatter is eliminated, and the simplest form of the transfer function is beneficial to accurately establish the relationship between the selection of impedance parameters and the properties of the robot and the environment, optimize the impedance parameters to eliminate the remaining inertial response of the jitter and improve the algorithm convergence speed.

[0021] An adaptive controller is established, and the simulation comparison of the adaptive impedance control and the traditional control effect is carried out through MATLAB / Simulink. Subsequently, the performance of the adaptive algorithm under different parameters and different environmental conditions is simulated and compared to reveal the influence of the external environment input on the optimal parameters.

[0022] A target function with the Pareto optimal solution as the goal is constructed, a dataset of the optimal parameters of the environment is obtained based on the meta-heuristic algorithm and combined with the BLS network to construct an optimal parameter optimizer, which further improves the stability and robustness of the system in the unstructured environment, and selects the current optimal parameters in real time for different environments to form an impedance control parameter optimization strategy.

[0023] (III) Beneficial effects

[0024] Compared with the prior art, the present invention provides an ultrasonic robot end zero-stiffness constant-force driver and its control method, having the following beneficial effects:

[0025] In the present invention, a negative stiffness with high-order non-linear characteristics is generated by the interaction of the moving magnet group and the fixed magnet through the magnetic field, and a linear positive stiffness is generated by the deformation of the linear spring. The resultant force of the two partially cancels each other out to achieve a small-amplitude wave force output with multi-stable characteristics. Compared with the traditional quasi-zero stiffness system with a monotonically increasing output force characteristic, the small-amplitude fluctuating output force of the multi-stable quasi-zero stiffness driver can achieve an approximate constant-force output with a larger stroke and a higher load; combined with the feedforward-feedback hybrid control algorithm, the small-amplitude fluctuation of the output force caused by the multi-stable characteristics is eliminated, and while achieving a large-stroke non-fluctuating constant-force output, the system exhibits a zero-stiffness characteristic independent of the feedback signal, improving the system robustness and environmental adaptability; the self-adaptability in the zero-stiffness constant-force interval reduces the requirements for the path planning of the ultrasonic robot, improves the force-position control efficiency, and ensures the safety of clinical operations and the image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic structural diagram of the first perspective of the ultrasonic robot end zero-stiffness constant-force driver of the present application;

[0027] Figure 2 It is a schematic internal structural diagram of the ultrasonic robot end zero-stiffness constant-force driver of the present application;

[0028] Figure 3 It is a schematic structural diagram of the top cover of the present application;

[0029] Figure 4 It is a schematic structural diagram of the base of the present application;

[0030] Figure 5 It is a schematic diagram of a high-linearity magnetic negative stiffness structure of the present application;

[0031] Figure 6 It is a force-displacement curve of the zero-stiffness generation principle of the present application;

[0032] Figure 7 It is a schematic diagram of the control method of position-based force tracking adaptive impedance of the present application;

[0033] Figure 8 It is a schematic diagram of the constant-force driving principle based on feedforward control of the present application;

[0034] Figure 9 It is a schematic diagram of the complete working scheme of the ultrasonic robot end zero-stiffness constant-force driver of the present application;

[0035] Figure 10 It is a schematic diagram of the composition of the ultrasonic robot system of the present application;

[0036] Figure 11 This is a schematic structural diagram of the second perspective of the zero-stiffness constant-force actuator at the end of the ultrasonic robot of the present application.

[0037] In the figure:

[0038] 1. Spring seat, 2. Linear spring, 3. Control circuit board, 4. Linear servo motor, 5. Displacement sensor, 6. Base, 8. Top cover, 9. Bearing seat, 10. Coil assembly, 11. Fixed magnet, 12. Moving magnet group, 13. Linear bearing, 14. Fixed magnet holder, 15. Optical axis, 17. Main frame, 18. 3D camera, 19. Ultrasonic probe, 20. Six-axis force sensor, 21. Six-axis robotic arm, 22. Zero-stiffness constant-force actuator at the end of the ultrasonic robot

[0039] 601. Displacement sensor slot, 602. First bearing hole, 603. Fixed magnet holder slot, 604. Second bearing hole, 605. Linear servo motor slot

[0040] 801. Spring seat slot, 802. First optical axis hole, 803. Second optical axis hole, 804. Moving magnet group assembly slot, 1201. Large spacer block, 1202. Small spacer block Detailed implementation manners

[0041] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0042] Please refer to Figures 1 to 11 , a zero-stiffness constant-force actuator at the end of an ultrasonic robot, including a housing, a positive stiffness structure, a negative stiffness structure, and a control circuit board 3.

[0043] The housing includes an external main frame, a top cover 8, a base 6, and internal bearings. The base 6 and the top cover 8 are connected by an optical axis 15 and a linear bearing 13.

[0044] Refer to Figure 3 , the top cover 8 includes a spring seat slot 801. The two optical axis holes include a first optical axis hole 802 and a second optical axis hole 803. The moving magnet group assembly slot 804. The spring seat 1 can move up and down under the control of the linear servo motor 4 through the spring seat slot 801. The moving magnet group 12 is fixed on the top cover 8 through the moving magnet group assembly slot 804. The optical axis 15 is fixed on the top cover 8 through the first optical axis hole 802 and the second optical axis hole 803.

[0045] Refer to Figure 4, the positive stiffness structure includes a linear servo motor 4, a displacement sensor 5, a linear spring 2 and a spring seat 1. The linear servo motor 4 is fixedly installed on a base 6. The spring seat 1 is fixedly installed on the output shaft of the linear servo motor 4. The linear spring 2 is arranged between a top cover 8 and the spring seat 1. The base 6 includes a displacement sensor groove 601, two bearing holes including a bearing hole one 602 and a bearing hole two 604, a fixed magnet holder groove 603, and a linear servo motor groove 605. The displacement sensor 5 is fixed on the base through the displacement sensor groove 601, and its upper end is in unconstrained contact with the top cover 8, thereby measuring the displacement of the upper part of the structure. A bearing seat 9 and a linear bearing 13 are fixed on the base 6 through the bearing hole one 602 and the bearing hole two 604. The linear servo motor 4 is fixed on the base 6 through the linear servo motor groove 605. A fixed magnet holder 14 is fixed on the base 6 through the fixed magnet holder groove 603. A fixed magnet 11 is fixed on the fixed magnet holder 14.

[0046] See Figure 5 , the magnetic negative stiffness structure includes a moving magnet group 12 and a fixed magnet 11. The moving magnet group 12 includes magnets and two different specifications of spacer blocks, including a large spacer block 1201 and a small spacer block 1202. Through the design of the structure and position, a high-nonlinear magnetic negative stiffness mechanism is formed, which can generate a large negative stiffness value, and at the same time has a good low-fluctuation approximate constant force output within a certain stroke range, and the structure is simple and easy to implement. A coil assembly 10 is added to the permanent magnet assembly. By changing the coil current, the electromagnetic force received by the coil assembly 10 in the magnetic field is controlled, so that different magnitudes of electromagnetic forces are provided at different working positions to cancel the magnetic force fluctuations in the working direction, thereby realizing the active trimming and regulation of the negative stiffness of the mechanism.

[0047] See Figure 6 , the force provided by the magnetic negative stiffness mechanism is as Figure 6 shown as "magnetic force" in the legend. The force provided by the linear spring 2 is as Figure 6 shown as "spring force" in the legend. The force obtained by the zero-stiffness constant force mechanism is as Figure 6 shown as "resultant force" in the legend. It can be seen from the figure that the curve of the legend "resultant force" is close to the target constant force in magnitude within the effective stroke range and crosses the target force axis multiple times. Compared with the traditional "resultant force" monotonically increasing quasi-zero stiffness mechanism, the near-constant force action interval of the "resultant force" is improved. Therefore, through the array arrangement of magnets, the quasi-zero stiffness characteristics of large stroke and low fluctuation can be achieved simultaneously.

[0048] See Figure 7The control part of the whole system is mainly divided into an inner loop of position control and an outer loop of impedance control. The inner loop control adopts the PID method. For the external input environment, the contact force between the ultrasonic probe 19 at the end of the robot and the medical dummy is obtained through the six-dimensional force sensor 20, and the force deviation is obtained by subtracting it from the expected force. After that, the force deviation is input into the impedance controller to generate a position correction amount, and the reference trajectory point of the trajectory planning is corrected to achieve the purpose of compliant control. The selection of impedance control parameters needs to consider the steady-state effect and the dynamic response of the system. Considering the feedforward-feedback hybrid control of the flexible drive, the system stiffness has a zero stiffness characteristic, and the transfer function has the simplest form. In order to prevent the sudden change of axial force in sudden situations, the proportional-differential control of the contact force is introduced into the impedance model, and the feedforward-feedback Under the influence of feedback hybrid control, the relationship between the selection of impedance parameters and the properties of the robot and the environment is analyzed. The impedance parameters are optimized to eliminate jitter and improve the convergence speed of the algorithm. An adaptive controller is established, and a simulation comparison of the adaptive impedance control and traditional control effects is performed through MATLAB / Simulink. Subsequently, the performance of the adaptive algorithm under different parameters and different environmental conditions is simulated and compared to reveal the influence of external environmental input on the optimal parameters. An objective function with Pareto optimal solution as the target is constructed. Based on the meta-heuristic algorithm, a data set of the optimal environmental parameters is obtained and an optimal parameter optimizer is constructed in combination with the BLS network, thereby further improving the stability and robustness of the system in unstructured environments. The current optimal parameters are selected in real time for different environments to form an impedance control parameter optimization strategy.

[0049] See Figure 8 The advantage of the large-stroke constant-force drive of the multi-stable system requires the use of position feedforward control based on electromagnetic actuators to offset the restoring force fluctuations between stable equilibrium positions. The fluctuation elimination mechanism is as follows: Figure 8 As shown in the figure, the dotted line represents the original restoring force curve, and the solid line represents the adjusted restoring force curve. After two-step adjustment, the driving scheme can set the contact force to the target constant force and effectively expand the action range of the constant contact force.

[0050] The present application is applied to an ultrasonic robot, including the ultrasonic robot end zero-stiffness constant force driver 22 in the above embodiment. The ultrasonic robot also includes a 3D camera 18, an ultrasonic probe 19, a six-dimensional force sensor 20, and a six-axis robotic arm 21.

[0051] Example 2, see Figure 9, this embodiment includes the zero-stiffness constant-force actuator at the end of the ultrasonic robot in the above embodiment. At the same time, this example demonstrates the complete working scheme of the zero-stiffness constant-force actuator at the end of the ultrasonic robot, adding a 3D camera 18, a main frame 17, a six-axis force sensor 20, and an ultrasonic probe 19. During the ultrasonic scanning process, it is necessary to use the 3D camera 18 to collect image depth data, and complete the preprocessing of the spatial point cloud data through four processing methods: downsampling, plane filtering, noise filtering, and normal estimation. From the preprocessed point cloud model, a set of path points containing spatial pose information is extracted, and then the path points are sent to the robot for trajectory planning, so as to realize the function of automatic ultrasonic scanning. The extraction method of the scanning path points is to extract all the point cloud information with normal vectors within a width range of 1 mm at intervals of 50 mm, and then use the spline interpolation method to smooth the path, and convert the point cloud path containing normal vector information on the scanning path into 5000 trajectory points, thus completing the trajectory planning. Subsequently, based on the bilinear discretization method, the adaptive control algorithm is converted from the Laplace domain to the discrete domain, and through appropriate initialization, the control equation is iteratively calculated to obtain the real-time position deviation of the desired force. Finally, based on Qt, the upper computer software is developed for data acquisition and processing, improving the development of the system communication module, controller module, upper computer display module, inverse kinematics module, and point cloud trajectory planning module, and on this basis, improving the control system of the prototype. Subsequently, a test experiment of the ultrasonic scanning robot experimental prototype is carried out on a medical phantom, and the six-axis force sensor 20 is used to record the change of the difference between the measured force and the desired force, and further optimize the impedance control parameter optimization strategy under various environments.

[0052] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An ultrasonic robot end zero-stiffness constant-force driver, which is applied to an ultrasonic robot, is characterized in that Comprising: A housing, the housing includes an external main frame, a top cover (8) and a base (6), and the base (6) and the top cover (8) are connected by an optical axis (15) and a linear bearing (13); A negative stiffness structure, the negative stiffness structure includes a fixed magnet (11) and a moving magnet group (12), the fixed magnet (11) is fixedly installed on the base (6), the moving magnet group (12) is fixedly installed on the top cover (8), the moving magnet group (12) includes magnets and two types of spacers, and the magnets and the spacers are alternately stacked; A positive stiffness structure, the positive stiffness structure includes a linear servo motor (4), a displacement sensor (5), a linear spring (2) and a spring seat (1), the linear servo motor (4) is fixedly installed on the base (6), the spring seat (1) is fixedly installed on the output shaft of the linear servo motor (4), and the linear spring (2) is arranged between the top cover (8) and the spring seat (1); The moving magnet group (12) and the fixed magnet (11) interact through a magnetic field to generate a high-order nonlinear negative stiffness, and the linear spring (2) generates a linear positive stiffness through deformation. The combined force of the two partially cancels out to achieve a quasi-zero stiffness approximate constant force output with multi-stable characteristics; A control circuit board (3), the control circuit board (3) is installed in the base (6), and the control circuit board (3) is electrically connected to the linear servo motor (4) and the displacement sensor (5).

2. The ultrasonic robot end zero-stiffness constant force driver according to claim 1, characterized in that The misaligned configuration of the magnets and the spacers in the moving magnet group (12) forms a stator-rotor misaligned structure. The spacers include a large spacer (1201) and a small spacer (1202), and the high-order nonlinear characteristics of the magnetic negative stiffness are realized through alternating arrangement.

3. The ultrasonic robot end zero-stiffness constant-force driver according to claim 2, characterized in that, The displacement sensor (5) is fixedly installed on the base (6), and the sensor rod of the displacement sensor (5) is in unconstrained contact with the top cover (8) for real-time measurement of the displacement of the top cover (8).

4. An ultrasonic robot uses an end zero-stiffness constant-force driver of the ultrasonic robot described in claim 3, and is characterized in that, Further comprising: A six-axis robotic arm (21), and the ultrasonic robot end zero-stiffness constant force driver is installed at the end of the six-axis robotic arm (21); An ultrasonic probe (19), a six-axis force sensor (20) and a 3D camera (18), and the ultrasonic probe (19), the six-axis force sensor (20) and the 3D camera (18) are integrated on the ultrasonic robot end zero-stiffness constant force driver; A host computer system, the host computer system is used to realize data acquisition, trajectory planning and real-time operation of an adaptive control algorithm. The driver adjusts the restoring force curve through a feedforward control mechanism and combines an electromagnetic actuator to eliminate the multi-stable output force fluctuation and achieve a large-stroke zero-stiffness constant force output.

5. An ultrasonic robot according to claim 4, wherein The host computer system is developed based on Qt and includes a communication module, a controller module, a point cloud trajectory planning module and an inverse kinematics module.

6. An ultrasonic robot according to claim 5, wherein, The ultrasonic robot end zero-stiffness constant force driver is tested on a medical mannequin. The deviation between the actual output force and the desired force is calculated in real time through the data of the six-axis force sensor (20) and the displacement sensor (5), and the impedance control parameters are dynamically optimized.

7. A control method for an ultrasonic robot end zero-stiffness constant-force driver, which uses an ultrasonic robot described in claim 6. Based on the force-displacement curve characteristics, the nonlinear deviation between the magnetic negative stiffness and the spring positive stiffness is compensated by feedforward control. It is characterized in that Including the following steps: S1. Obtain the real-time output force of the six-axis robotic arm (21) on the target contact surface through the six-axis force sensor (20). At the same time, obtain the inertial forces of the ultrasonic probe (19) and the zero-stiffness constant-force driver (22) at the end of the ultrasonic robot by taking the second derivative of the data of the displacement sensor (5), and calculate the deviation from the desired contact force by summation; S2. Input the force deviation into the impedance controller to generate a position correction amount and correct the reference trajectory points of the trajectory planning; S3. Adopt a feedforward-feedback hybrid control algorithm. The inner loop is a position control based on PID, and the outer loop is an impedance control based on proportional-derivative control. Among them, the impedance parameters are optimized by a metaheuristic algorithm, and an adaptive parameter optimizer is constructed by combining with the BLS network; S4. Verify the dynamic responses of the adaptive impedance control and the traditional control through MATLAB / Simulink simulation, and optimize the parameters to eliminate system jitter and improve the convergence speed.

8. A control method for an ultrasonic robot end zero-stiffness constant-force driver according to claim 7, characterized in that After filtering the displacement of the sudden axial force by using the dynamic adjustment effect in the constant-force interval with zero-stiffness characteristics, introduce the proportional-derivative control of the contact force into the impedance control model to suppress the sudden change of the axial force. And the selection of the impedance parameters is based on the Pareto optimal solution objective function, and the optimal parameters are dynamically adjusted through the real-time environment input.

9. A control method for an ultrasonic robot end zero-stiffness constant force driver according to claim 8, characterized in that, The trajectory planning is realized through the following steps: S1. Use the 3D camera (18) to collect the target point cloud data, and extract the path points including the normal vector after preprocessing; S2. Extract the path points within a range of 1 mm width at intervals of 50 mm, and generate 5000 smooth trajectory points by using the spline interpolation method; S3. Combine the bilinear discretization method to transform the adaptive control algorithm from the Laplace domain to the discrete domain and calculate the position deviation in real time.

Images