Flexible clamping mechanism and clamp self-adaptive adaptation method and device

Abstract

The invention relates to a flexible clamping mechanism and clamp self-adaptive adaptation method and device, the self-adaptive adaptation device comprises an assembly robot and a pose acquisition module, and the flexible clamping mechanism is carried at the tail end of the assembly robot; the pose acquisition module is used for acquiring a first relative pose error between the compliant clamping mechanism in an initial state and the clamp module for pre-alignment, and acquiring a second relative pose error between the compliant clamping mechanism and the clamp module after pre-alignment for self-adaptive adaptation; the assembly robot is used for driving the flexible clamping mechanism to correspondingly act according to the first relative pose error so as to realize pre-alignment between the flexible clamping mechanism and the clamp module; and then according to the second relative pose error, the compliant clamping mechanism is driven to achieve self-adaptive adaptation between the compliant clamping mechanism and the clamp module through controlled micro-amplitude vibration in combination with progressive feeding motion. Self-adaptive adaptation can be achieved without depending on a force sensor, and excessive deformation and structural damage of parts are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of micro-assembly, and in particular to a method and a device for adaptively matching a compliant clamping mechanism with a clamp. Background Art

[0002] Microassembly technology is a precision manufacturing process, the core of which is to achieve precise assembly of tiny parts ranging from micrometers to millimeters within the micron-level error range, thereby building microsystems with specific functions.

[0003] The key technical links in the microassembly process include precise positioning, reliable clamping and controlled release of microdevices. In the actual assembly process, especially when facing the need to assemble cross-scale devices at the same workstation, the traditional method usually uses a clamping mechanism of a specific size to perform the operation. However, this method has obvious limitations: on the one hand, frequently replacing the clamping mechanism to match the device size will significantly reduce the assembly efficiency; on the other hand, repeated disassembly and assembly processes may introduce additional assembly errors, thereby affecting the quality and performance reliability of the final product. To this end, Wang Rixin et al. from South China University of Technology designed a microgripper with replaceable modular tools for cross-scale microassembly. By replacing different clamping modules, it can adapt to the assembly requirements of different sizes (Wang R, Zhang X, Zhu B, et al. A Topology-Optimized Compliant Microgripper With Replaceable Modular Tools for Cross-Scale Microassembly[J]. IEEE / ASME Transactions on Mechatronics, 2024, 29(3): 2067-2078).

[0004] However, in order to achieve universal adaptability between the clamping mechanism and different fixture modules and ensure operational stability during the clamping and releasing process, the system design adopts an interference fit method. While this fit method improves connection reliability, it also brings the risk of jamming during assembly, seriously affecting the assembly success rate. To address this technical difficulty, existing studies have mostly adopted force / torque sensing technology to judge the assembly status and make dynamic adjustments by real-time monitoring of contact force, thereby improving assembly efficiency. For example, Dong et al. added force sensors to each joint of the robot and used the force feedback information of each joint to achieve flexible control of the end, completing high-precision axial hole assembly. (Dong L, Ma J, Cao J, et al. Serial–parallel cooperative assembly approach for precision micro-assembly of axial holes[J]. Mechanical Sciences, 2024, 15(2): 653-665). The core of the above technology relies on multi-dimensional force / torque sensors, which not only significantly increases the system cost, but also restricts the selection range of robot drive motors due to the volume and weight limitations of the sensors. Secondly, this technology requires stable contact between the shaft and the hole to obtain force feedback information, making it more suitable for assembly scenarios with rigidly fixed parts. However, when faced with floating micro-flexible parts, this method is often difficult to implement effectively. Not only is the control effect poor, but improper contact force can also lead to excessive deformation of the part or even structural damage. Summary of the Invention

[0005] In response to the problems existing in the prior art, the purpose of the present invention is to provide a method and device for adaptively adapting a compliant clamping mechanism to a fixture, which can achieve adaptive adaptation of the clamping mechanism and the fixture without relying on a force sensor, thereby avoiding excessive deformation of parts and structural damage.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A compliant clamping mechanism and fixture adaptive adaptation device includes a frame;

[0008] The frame is equipped with an assembly robot and a posture acquisition module, and the end of the assembly robot is equipped with a flexible clamping mechanism;

[0009] The posture acquisition module is used to obtain a first relative posture error between the flexible clamping mechanism and the fixture module in an initial state for pre-alignment, and to obtain a second relative posture error between the flexible clamping mechanism and the fixture module after pre-alignment for adaptive adaptation;

[0010] The assembly robot is connected to the posture acquisition module, and is used to drive the flexible clamping mechanism to move accordingly according to the first relative posture error to achieve pre-alignment between the flexible clamping mechanism and the fixture module, and then drive the flexible clamping mechanism to achieve adaptive adaptation between the flexible clamping mechanism and the fixture module through controlled micro-vibration combined with progressive feed motion according to the second relative posture error.

[0011] Furthermore, the posture acquisition module includes a first industrial camera and a second industrial camera, and the first industrial camera and the second industrial camera are arranged corresponding to the fixture module respectively; the first industrial camera is used to obtain the posture error between the flexible clamping mechanism and the fixture module in the horizontal plane of space; the second industrial camera is used to obtain the posture error between the flexible clamping mechanism and the fixture module in the vertical plane of space.

[0012] Furthermore, the first industrial camera and the second industrial camera are arranged orthogonally, and are both equipped with a telecentric lens and an independent light source module.

[0013] Furthermore, the frame is provided with a macro motion platform, a camera translation module and a camera rotation module; the macro motion platform is connected to the assembly robot to drive the assembly robot to move over a large range; the camera translation module is connected to the first industrial camera to drive the first industrial camera to translate along the x and y directions; the camera rotation module is connected to the second industrial camera to drive the second industrial camera to rotate around the z axis.

[0014] Furthermore, the assembly robot is provided with linear motors that translate in different directions and rotary motors that rotate in different directions, and each joint has micro-movement and vibration functions.

[0015] Furthermore, the rotary joint of each rotary motor is controlled by a proportional-integral controller, and each linear motor is controlled by a stepper motion control with a closed loop.

[0016] A method for adaptively adapting a flexible clamping mechanism to a fixture comprises the following steps:

[0017] The posture acquisition module acquires a first relative posture error between the compliant clamping mechanism and the fixture module in an initial state;

[0018] The assembly robot drives the compliant clamping mechanism to move accordingly according to the first relative posture error, thereby achieving pre-alignment between the compliant clamping mechanism and the fixture module;

[0019] The posture acquisition module acquires a second relative posture error between the pre-aligned compliant clamping mechanism and the fixture module;

[0020] The assembly robot drives the compliant clamping mechanism to achieve adaptive adaptation between the compliant clamping mechanism and the fixture module through controlled micro-vibration combined with progressive feed motion according to the second relative posture error.

[0021] The pre-aligned controlled micro-vibration combined with progressive feed motion is implemented by adding the vibration offset of each joint motion axis of the assembly robot to the feed motion trajectory, so that the flexible clamping mechanism feeds and vibrates simultaneously toward the fixture module.

[0022] Furthermore, the simultaneous feeding and vibrating of the compliant clamping mechanism comprises the following steps:

[0023] Obtaining a relative error Δx in the feed direction between the compliant clamping mechanism and the fixture module according to the second relative posture error;

[0024] According to the relative error in the feed direction and the given feed speed v x Set the motion law of the feed direction to x(t) = x0 + v x t,t∈[0,Δx / v x ];

[0025] According to the given vibration amplitude a and vibration frequency f, the motion law of the vibration direction is set to z(t)=z0+a sin(2πft);

[0026] At time t∈[0,Δx / v x ] According to the motion laws in the feeding and vibration directions and the inverse solution of the robot kinematics, the joints of the assembly robot are driven to move accordingly, thereby realizing the simultaneous feeding and vibration of the flexible clamping mechanism.

[0027] Furthermore, at time t = Δx / v x When the relative error is less than the given allowable value, the adaptive adaptation ends. Otherwise, the feeding and vibration are repeated according to the relative error until the relative error is less than the given allowable value or the iteration limit is reached.

[0028] In general, the present invention has the following advantages:

[0029] (1) The present invention proposes a vision-guided vibration adaptation solution that can achieve the adaptive adaptation process of the flexible clamping mechanism and the fixture without the need for force sensors and complex algorithms.

[0030] (2) The vibration frequency of the present invention is relatively low, and vibration can be achieved by relying on robot joint control, thereby eliminating the need for additional equipment to stimulate vibration, thereby reducing the cost of the equipment and the volume of the overall system.

[0031] (3) The visual algorithm proposed in the present invention can successfully complete the detection even when the fixture module is slightly deformed and blocked to a certain extent, thereby improving the robustness of the system.

[0032] (4) The adaptation strategy proposed in the present invention only calls the visual algorithm to detect errors before and after the robot moves. Compared with real-time detection, it not only improves the success rate of detection, but also reduces the occupation of computer system resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Schematic diagram of the overall structure of the device of the present invention.

[0034] Figure 2 It is a structural schematic diagram of the assembly robot and the end flexible clamping mechanism in the present invention.

[0035] Figure 3 This is a flow chart of the adaptive vibration adaptation method of the present invention.

[0036] Figure 4 for Figure 1 A is an enlarged schematic diagram.

[0037] In the picture:

[0038] 1-Frame, 2-Camera rotation module, 3-Second industrial camera, 4-Telecentric lens, 5-Camera translation module, 6-First industrial camera, 7-Light source module, 8-Macro motion platform, 9-Assembly robot, 10-Flexible clamping mechanism, 11-Fixture library, 12-Fixture module, 13-Z-axis rotation motor, 14-Y-axis linear motor, 15-X-axis linear motor, 16-Z-axis linear motor, 17-Y-axis rotation motor, 18-End gripper. DETAILED DESCRIPTION

[0039] The present invention will be described in further detail below.

[0040] like Figure 1 、 Figure 4 As shown, a flexible clamping mechanism and fixture adaptive adaptation device includes a frame 1, an assembly robot 9 and a posture acquisition module are provided on the frame 1, and a flexible clamping mechanism 10 is mounted on the end of the assembly robot 9;

[0041] The posture acquisition module is used to obtain a first relative posture error between the flexible clamping mechanism 10 and the clamping module 12 in an initial state for pre-alignment, and is used to obtain a second relative posture error between the flexible clamping mechanism 10 and the clamping module 12 after pre-alignment for adaptive adaptation;

[0042] The assembly robot 9 is connected to the posture acquisition module, and is used to drive the flexible clamping mechanism 10 to move accordingly according to the first relative posture error to achieve pre-alignment between the flexible clamping mechanism 10 and the clamping module 12, and then drive the flexible clamping mechanism 10 to achieve adaptive adaptation between the flexible clamping mechanism 10 and the clamping module 12 through controlled micro-vibration combined with progressive feed motion according to the second relative posture error.

[0043] Specifically, the adaptive adaptation device includes a camera rotation module 2 at the bottom of the rack 1, a second industrial camera 3, a telecentric lens 4, a camera translation module 5 at the top of the rack 1, a first industrial camera 6, a light source module 7, a macro motion platform 8, an assembly robot 9, a flexible clamping mechanism 10, a fixture library 11, and fixture modules 12 of different sizes.

[0044] The fixture library 11 is located on the center platform of the frame 1 and is connected by bolts. The slotted design of the connection holes facilitates subsequent fine-tuning of the relative positions of the fixture library 11, the second industrial camera 3, the first industrial camera 6, and the assembly robot 9. The fixture library 11 is designed with slots of different sizes, with slots of different sizes accommodating fixture modules 12 of different sizes. Each slot has a unique identifier above for visual inspection and a small stop below to facilitate the removal and placement of the fixture module 12 by the flexible clamping mechanism 10.

[0045] The posture acquisition module includes a second industrial camera 3 and a first industrial camera 6. The second industrial camera 3 is located behind the fixture library 11 and is connected to the bottom camera rotation module 2, allowing the camera to rotate around the z-axis. The center height of its field of view is slightly higher than the fixture library 11. It is mainly used to obtain the posture information of the flexible clamping mechanism 10 and the fixture module 12 on the vertical plane of space;

[0046] The first industrial camera 6 is located above the fixture library 11 and is connected to the camera translation module 5 at the top. Combined with the offset of the camera translation module 5 and the mounting position of the rack 1, the first industrial camera 6 can achieve a certain distance of x and y movement. The center of its field of view is close to the center of the fixture library 11. It is mainly used to obtain the position information of the flexible clamping mechanism 10 and the fixture module 12 on the horizontal plane of space;

[0047] The assembly robot 9 is located in front of the fixture library 11 and is connected to the macro-motion platform 8 on top. The macro-motion platform 8 has a travel range of 1 meter, allowing the assembler to replace the fixture module 12 and then cooperate with other platforms. The end of the assembly robot 9 is equipped with a flexible clamping mechanism 10. In this embodiment, the flexible clamping mechanism 10 has an end gripper 18 driven by two piezoelectric ceramics, which allows the jaws to clamp over a certain range of travel.

[0048] like Figure 2 As shown, the assembly robot 9 in this embodiment is a five-degree-of-freedom robot, including two rotary motors and three linear motors, which are, from top to bottom, a rotary motor 13 around the z-axis, a linear motor 14 around the y-axis, a linear motor 14 around the x-axis, a linear motor 15 around the z-axis, 16 and a rotary motor 17 around the y-axis. The stroke of the rotary motors is ±π, and the stroke of the linear motors is ±15mm. Each joint can achieve high-precision micro-movement and vibration.

[0049] In this embodiment, the first industrial camera 6 and the second industrial camera 3 are positioned orthogonally and each is equipped with a telecentric lens 4 with a resolution of 2592×2048 and a field of view of 2.4×2.3 mm. Therefore, the relative positions of the first industrial camera 6, the second industrial camera 3, the assembly robot 9, and the fixture library 11 must be adjusted so that the fixture modules 12 in the fixture library 11 and the flexible clamping mechanism 10 at the end of the assembly robot 9 appear within the industrial camera's field of view and are relatively centrally located to facilitate subsequent data collection. Furthermore, each industrial camera is equipped with an independent light source module 7 to provide illumination.

[0050] like Figure 3 As shown, the present invention provides a method for adaptively matching a compliant clamping mechanism with a clamp, comprising the following steps:

[0051] S1. System initialization, including initialization of the motor, initialization of the camera, selection of the matching object, loading of the matching template, selection of the ROI area, setting the upper limit of the number of iterations for a single match, and the allowable value of the matching error. The initialization of the motor includes the selection of the communication port and the setting of the PI closed-loop parameters. The initialization of the camera includes the configuration of the exposure time and sampling frequency, etc. In this embodiment, each linear motor uses a relative encoder. The absolute zero point of each motor needs to be determined in advance, and its communication port and PI closed-loop parameters need to be configured. Configure the internal and external parameters of the camera, adjust the camera exposure time, sampling frequency and light source brightness until each component can be clearly identified in the camera field of view. Set the upper limit of the number of iterations for a single match to 10 times, the allowable value of the pre-alignment error to 0.05mm, and the allowable value of the adaptive adaptation error to 0.02mm. Set the fixture module 12 selected in this adaptation task, accurately locate the slot position of the module in the image obtained by the first industrial camera 6 according to the identifier set in advance, record the ROI area information, and perform a brief search to determine the authenticity and approximate position of the module in the fixture library 11, so as to facilitate the improvement of subsequent algorithm retrieval efficiency.

[0052] In order to ensure the real-time performance of visual inspection, the templates for visual matching are prepared in advance and loaded into the program during the initialization phase, including the slot position identifier, SIFT matching template, and the left and right sub-templates of the template matching algorithm based on edge detection. Among them, the slot position identifier is mainly used to segment each slot in the image, reduce the amount of matching data, improve matching efficiency, and also prevent mutual interference of information between different fixture modules 12; the SIFT matching template is mainly used to achieve real-time tracking and positioning of the end of the fixture module 12, and obtain the relative posture error between the end of the flexible clamping mechanism 10 and the target pre-alignment position of the fixture module 12. Considering that the fixture module 12 has the characteristics of less texture and easy deformation, in this embodiment, the obtained top-view grayscale image of the fixture module 12 is divided into rigid areas and flexible areas according to the degree of deformation, and the image is pre-processed and divided into 50×50 pixel blocks. The tensor eigenvalue and anisotropy of each block are calculated using the following formula:

[0053]

[0054] Among them, S xx 、S yy 、S xy Respectively represent the second-order derivatives of the image in the x and y directions, as well as the second-order mixed partial derivatives in the xy directions. max ,λ min is the extreme value of the eigenvalue of the image structure tensor. A represents the anisotropy of the image block. The structure tensor can well reflect the grayscale gradient distribution in the image, while the eigenvalue of the tensor can highlight the strength of the gradient change. In combination with the anisotropy of the tensor, an appropriately sized region is selected to extract edge information as the recognition template for the fixture module 12. Considering the symmetry of the fixture module 12, in this embodiment, a 50×200 pixel region is selected on each side of the fixture module 12 as a sub-template.

[0055] S2. Use the SDK provided by the camera to obtain images from the second industrial camera 3 and the first industrial camera 6. Use the obtained images as input to the visual inspection algorithm. Use the image processing algorithm to obtain the position and rotation error of the end of the flexible clamping mechanism 10 and the entrance of the fixture module 12 in the xy plane from the image of the first industrial camera 6. Obtain the relative position error of the z axis from the image of the second industrial camera 3, thereby obtaining the first relative posture error.

[0056] S3. Based on the joint space measurements obtained by the current joint position sensors of the assembly robot 9, the current task space measurements at the end of the compliant clamping mechanism 10 are calculated using a forward kinematic solution. Based on the current task space position of the compliant clamping mechanism 10 and the task space error between the compliant clamping mechanism 10 and the entrance of the fixture module 12, the target position of the compliant clamping mechanism 10 in the task space is calculated. The target motion position of each joint of the assembly robot 9 is then determined using an inverse kinematic solution. The robot controller is used to cause each joint to move in a closed-loop manner to the target motion position.

[0057] In this embodiment, the end of the assembly robot 9 is controlled by a proportional controller with visual error as feedback. The rotary joints of the assembly robot 9 are controlled by a proportional-integral controller, and the linear joints of the assembly robot 9 are controlled by a closed-loop stepper motion control.

[0058] The image of the first industrial camera 6 and the position error of the xy plane are obtained again, and the position of the end of the flexible clamping mechanism 10 is fine-tuned according to the error until the error is less than the pre-alignment error tolerance or the number of iterations reaches the upper limit. It is then considered that the flexible clamping mechanism 10 has reached the entrance position of the adaptation fixture module 12, and a visually guided vibration adaptation process can be performed to achieve adaptive adaptation.

[0059] S4. A template matching algorithm based on edge detection is used to obtain a relative posture error between the end of the compliant clamping mechanism 10 and the target matching position of the clamp module 12, thereby obtaining a second relative posture error.

[0060] S4.1. Select a specific ROI area in the image acquired by the camera according to the selected matching object.

[0061] S4.2. Obtain the position of the fixture module 12 in the image based on the left and right templates, and determine the degree of deformation of the fixture module 12 based on the recognition effect: if the deformation is small, repair the recognition result and calculate the target matching position of the fixture module 12; if the deformation is large, enter the robot correction program.

[0062] S4.3. The matching algorithm is the NCC algorithm, which is expressed as follows:

[0063]

[0064] in, represents the gradient of the template, Represents the gradient of the target matching area. The gradient information of each edge point in the template and the gradient information of the edge point in the target area are used as input to calculate the matching degree between the two.

[0065] S4.4. Record the positioning results obtained this time to narrow the search area for next identification.

[0066] S5. Plan the motion trajectory according to the error and the given speed, and add the vibration offset at each moment to the motion trajectory of the corresponding motion axis of the assembly robot 9 according to the given vibration law. When the target matching position error between the end of the flexible clamping mechanism 10 and the clamping module 12 is less than the given allowable value or the number of closed-loop control reaches the upper limit, the adaptive adaptation ends.

[0067] S5.1. Preset the feed speed V of the assembly robot 9 x , vibration frequency f and vibration amplitude a. In this embodiment, the feed speed is 0.5 mm / s, the vibration frequency is 20 Hz, and the vibration amplitude is 0.02 mm. The selection of parameters is obtained by experiments. The actual values given here are for reference only. The specific settings need to refer to the actual situation, such as motor selection, accuracy requirements, etc.

[0068] S5.2. Obtain the top camera image and the position and rotational error of the end of the compliant gripping mechanism 10 relative to the target fixture fit position in the xy plane through the above process. The current position (x0, y0, z0) of the compliant gripping mechanism 10 is obtained based on the joint sensors and the forward kinematic solution. In this embodiment, the forward and inverse solutions and differential kinematics of the assembly robot 9 are obtained using the DH method.

[0069] S5.3, according to the feed direction error dx and the preset feed speed V x Calculate the time t of this vibration adaptation process f =dx / v x , set the x-direction trajectory to x(t)=x0+v x t,t∈[0,Δx / v x ], the motion trajectory in the vibration direction z is z(t) = z0 + a sin(2πft). Then, a timer is started from zero, and the assembly robot 9 is controlled to move according to the current time and the trajectory in each direction until the time is greater than the adaptation time t f .

[0070] S5.4: The motor controller plans a trajectory based on this error and a predefined vibration strategy, executing it without vision assistance. Upon reaching the trajectory endpoint, the vision system rechecks the error. If the error meets the requirements, the assembly is considered complete; otherwise, the process repeats. This approach reduces the frequency of time-consuming template matching, thereby improving real-time performance.

[0071] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A flexible clamping mechanism and a fixture adaptive adapter device, characterized by: Including rack; The frame is equipped with an assembly robot and a posture acquisition module, and the end of the assembly robot is equipped with a flexible clamping mechanism; The posture acquisition module is used to obtain a first relative posture error between the flexible clamping mechanism and the fixture module in an initial state for pre-alignment, and to obtain a second relative posture error between the flexible clamping mechanism and the fixture module after pre-alignment for adaptive adaptation; The assembly robot is connected to the posture acquisition module, and is used to drive the flexible clamping mechanism to move accordingly according to the first relative posture error to achieve pre-alignment between the flexible clamping mechanism and the fixture module, and then drive the flexible clamping mechanism to achieve adaptive adaptation between the flexible clamping mechanism and the fixture module through controlled micro-vibration combined with progressive feed motion according to the second relative posture error.

2. The adaptive fitting device according to claim 1, characterized in that: The posture acquisition module includes a first industrial camera and a second industrial camera, which are arranged corresponding to the fixture module respectively; the first industrial camera is used to obtain the posture error between the flexible clamping mechanism and the fixture module on the horizontal plane of space; the second industrial camera is used to obtain the posture error between the flexible clamping mechanism and the fixture module on the vertical plane of space.

3. The adaptive fitting device according to claim 2, characterized in that: The first industrial camera and the second industrial camera are arranged orthogonally and are both equipped with a telecentric lens and an independent light source module.

4. The adaptive fitting device according to claim 2, wherein: The frame is equipped with a macro motion platform, a camera translation module, and a camera rotation module; the macro motion platform is connected to the assembly robot to drive the assembly robot to move over a large range; the camera translation module is connected to the first industrial camera to drive the first industrial camera to translate along the x and y directions; The camera rotation module is connected to the second industrial camera to drive the second industrial camera to rotate around the z-axis.

5. The adaptive fitting device according to claim 1, characterized in that: The assembly robot is equipped with linear motors that translate in different directions and rotary motors that rotate in different directions, and each joint has micro-movement and vibration functions.

6. The adaptive fitting device according to claim 5, characterized in that: The rotary joint of each rotary motor is controlled using a proportional-integral controller, and each linear motor uses stepper motion control with closed loop.

7. A method for adaptively adapting a compliant clamping mechanism to a fixture according to any one of claims 1 to 6, characterized in that: The following steps are included: The posture acquisition module acquires a first relative posture error between the compliant clamping mechanism and the fixture module in an initial state; The assembly robot drives the compliant clamping mechanism to move accordingly according to the first relative posture error, thereby achieving pre-alignment between the compliant clamping mechanism and the fixture module; The posture acquisition module acquires a second relative posture error between the pre-aligned compliant clamping mechanism and the fixture module; The assembly robot drives the compliant clamping mechanism to achieve adaptive adaptation between the compliant clamping mechanism and the fixture module through controlled micro-vibration combined with progressive feed motion according to the second relative posture error.

8. The method according to claim 7, wherein: The controlled micro-vibration combined with progressive feed motion is achieved by adding the vibration offset of each joint motion axis of the assembly robot to the feed motion trajectory, so that the flexible clamping mechanism feeds and vibrates simultaneously toward the fixture module.

9. The method according to claim 8, characterized in that: The simultaneous feeding and vibration of the compliant clamping mechanism includes the following steps: Obtaining a relative error Δx in the feed direction between the compliant clamping mechanism and the fixture module according to the second relative posture error; According to the relative error in the feed direction and the given feed speed v x Set the motion law of the feed direction to x(t) = x0 + v x t,t∈[0,Δx / v x ]; According to the given vibration amplitude a and vibration frequency f, the motion law of the vibration direction is set to z(t)=z0+a sin(2πft); At time t∈[0,Δx / v x ] According to the motion laws in the feeding and vibration directions and the inverse solution of the robot kinematics, the joints of the assembly robot are driven to move accordingly, thereby realizing the simultaneous feeding and vibration of the flexible clamping mechanism.

10. The method according to claim 9, characterized in that: At time t = Δx / v x When the relative error is less than the given allowable value, the adaptive adaptation ends. Otherwise, the feeding and vibration are repeated according to the relative error until the relative error is less than the given allowable value or the iteration limit is reached.

Images