Industrial robot collaborative compliant assembly vibration suppression device and method

Abstract

This invention provides a vibration suppression device and method for compliant assembly in industrial robot human-robot collaboration, comprising an industrial robot, a six-dimensional force sensor, a binocular vision unit, and a robot compliant control model. This invention proposes an adaptive moving average filtering vibration suppression method based on the robot system's operating state. By sensing the system's stable operation and continuously changing states, the size of the moving average filtering window is adjusted. This ensures both rapid response to external force information during stable system operation and avoids rapid changes in force and displacement correction information caused by continuous and rapid changes in the system's state, thus achieving vibration suppression through force-position coupling control.

Description

Technical Field

[0001] This invention belongs to the field of industrial robot technology, specifically relating to a vibration damping device and method for compliant assembly of industrial robots in human-machine collaboration. Background Technology

[0002] Large aerospace components are characterized by high assembly precision, high flexibility, and large weight. Traditionally, assembly is achieved using cranes and manual labor, but this method requires high operational skills, has low assembly efficiency, and is difficult to control in terms of positioning accuracy. Therefore, current research is exploring the use of industrial robots with force sensors to create a human-robot collaborative compliant assembly system. The robot can offset the load on the large components, and the force sensor senses the human's intentions to achieve compliant dragging assembly. This method can replace traditional hoisting installation methods. However, this method heavily relies on the stability of the robot's compliant control data. Fluctuations in this data can easily cause self-excited vibrations, damaging the robot's intelligent equipment, the workpiece, or even jeopardizing safety.

[0003] The assembly of large aerospace components by industrial robots requires high precision. However, issues such as large center of gravity offset of the end product, easy fluctuation of force sensor data, and variable external environmental conditions can easily lead to vibration during the human-robot collaboration process, affecting assembly stability and safety. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a vibration damping device and method for compliant assembly of industrial robots in human-robot collaboration. The solution of this invention can solve the problems existing in the prior art.

[0005] The technical solution of this invention:

[0006] According to the first aspect, a compliant assembly vibration damping device for industrial robots is provided, comprising an industrial robot, a six-dimensional force sensor, a binocular vision unit, and a robot compliant control model. The industrial robot model meets the requirements of the component to be installed. The six-dimensional force sensor is installed on the flange of the industrial robot and connected to the component to be installed via a mechanical connector. Several visual target points are set on the component to be installed. The six-dimensional force sensor acquires the force situation of the component to be installed and transmits it to the robot compliant control model. The binocular vision unit acquires the position of the visual target points and, based on the position of the visual target points, acquires the motion state of the component and sends it to the robot compliant control model. The robot compliant control model processes the acquired force situation, acquires the correction information of the industrial robot displacement, sends it to the industrial robot, and acquires the robot's corrected Cartesian space pose information. The Cartesian space pose information is combined with the component motion state acquired by the binocular vision unit to obtain the stability of the robot system. Based on the stability, the robot compliant control model is adjusted to achieve the vibration damping function of the system.

[0007] Furthermore, the robot compliance control model includes an adaptive moving average filtering unit, a force-position coupling control model, and a moving average filtering unit. The adaptive moving average filtering unit receives information from a six-dimensional force sensor and performs moving average filtering on multiple sets of sensor data, using the following formula: ,in This is the nth set of force data input into the force-potential coupling control model after filtering. To determine the window size for adaptive moving average filtering, F represents the force sensor data before filtering. The force data obtained after moving average filtering is input into the force-position coupling control model. This model calculates and outputs displacement correction information for the robot in six directions based on the obtained force data. This displacement correction information is then input into the moving average filtering unit, which performs moving average filtering on the obtained displacement correction information using the following formula: In the formula, The nth set of displacement correction information is filtered, and D is the displacement correction information output by the force-position coupling control. The filtered displacement correction information is sent to the industrial robot.

[0008] Furthermore, the window of the adaptive moving mean filter... The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, .

[0009] Furthermore, the force sensor data input into the force-position coupling control model is used for gravity compensation, zero drift compensation, and temperature drift compensation.

[0010] Furthermore, the motion state of the component acquired by the binocular vision unit is subjected to sliding mean filtering, as shown in the formula: In the formula, This is the robot end-effector pose information acquired by binocular vision after filtering. Where N is the sampling frequency of the binocular vision system, M is the robot end-effector pose information acquired by the binocular vision system before filtering, and N is the sampling frequency of the binocular vision system. M The size of the adaptive moving mean filter window for visual information.

[0011] Furthermore, the criteria for judging the stability of a system are as follows:

[0012] If the robot's pose information in Cartesian space changes continuously and stably, and the pose information of the end effector monitored by binocular vision changes continuously and stably, then the system will operate stably.

[0013] If the robot's pose information in Cartesian space changes continuously and stably, while the pose information of the end effector changes rapidly and repeatedly, the system will operate stably.

[0014] If the robot's pose information in Cartesian space changes rapidly and repeatedly, while the pose information of the binocular vision recognition and monitoring end effector changes continuously and stably, then the system is unstable.

[0015] If the robot's pose information in Cartesian space changes rapidly and repeatedly, and the pose information of the binocular vision recognition monitoring end effector changes rapidly and repeatedly, the system will be unstable.

[0016] According to the second aspect, a method for vibration suppression in compliant assembly using industrial robots in human-robot collaboration is provided, comprising the following steps:

[0017] Step 1: Obtain the force information of the workpiece to be assembled;

[0018] Step two: Apply adaptive moving average filtering to the force information;

[0019] The filtered force information is used to calculate the displacement correction information of the industrial robot;

[0020] The displacement correction information, after being filtered by the moving average, is sent to the industrial robot to control its movement.

[0021] Step 3: Obtain the Cartesian space pose information of the industrial robot;

[0022] Obtain the end-effector pose information of the industrial robot based on binocular vision;

[0023] Step 4: Based on the Cartesian space pose information and end-effector pose information obtained in Step 1, determine whether the industrial robot system is operating stably. If yes, proceed to Step 1; otherwise, proceed to Step 2 to correct the filtering parameters.

[0024] Furthermore, in step two, the formula for adaptive moving mean filtering is: ,in This is the nth set of force data after filtering. The window size for adaptive moving average filtering is denoted by F, where F represents the force sensor data before filtering.

[0025] Furthermore, the window of the adaptive moving mean filter... The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, .

[0026] The beneficial effects of this invention compared to the prior art are as follows:

[0027] 1) This invention proposes an adaptive sliding mean filtering vibration suppression method based on the operating state of a robot system. By sensing the stable operation and continuously changing state of the system, the size of the sliding mean filtering window is adjusted, which not only ensures the rapid response to external force information during the stable operation of the system, but also avoids the rapid changes in force information and displacement correction information caused by the continuous and rapid changes in the system's state, thus realizing vibration suppression through force-position coupling control.

[0028] 2) This invention employs a dual-layer adaptive sliding mean filter that combines force sensor-acquired information with force-position coupling control output displacement correction information. This reduces rapid changes in information from both the input and output perspectives, thereby reducing the likelihood of vibration during the robot's force-position coupling control process.

[0029] 3) This invention introduces system status monitoring based on multi-source information such as force sensor information and visual tracking information, which improves the accuracy of system status perception, avoids the instability or error of information that may exist in single perception information, and further ensures the vibration suppression effect of force-position coupling control of the system. Attached Figure Description

[0030] The accompanying drawings, which form part of this specification, are provided to further illustrate embodiments of the invention and, together with the textual description, explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0031] Figure 1 A schematic diagram of a vibration damping device for human-machine collaborative compliant assembly of an industrial robot, provided according to an embodiment of the present invention, is shown.

[0032] Figure 2 A flowchart of a human-machine force-position coupling control for an industrial robot according to an embodiment of the present invention is shown.

[0033] Figure 3 A schematic diagram of a human-machine collaborative compliant assembly vibration suppression method for industrial robots, provided by an embodiment of the present invention, is shown.

[0034] The above figures include the following reference numerals:

[0035] 1. Industrial robot; 2. Components to be installed; 3. Six-dimensional force sensor; 4. Target point; 5. Binocular vision unit. Detailed Implementation

[0036] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0038] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0039] like Figure 1As shown in the first aspect of the present invention, an industrial robot human-machine collaborative compliant assembly vibration suppression device is provided, including an industrial robot, a six-dimensional force sensor, a binocular vision unit, and a robot compliant control model. The industrial robot model meets the requirements of the component to be installed. The six-dimensional force sensor is installed on the flange of the industrial robot and connected to the component to be installed through a mechanical connector. Several visual target points are set on the component to be installed. The six-dimensional force sensor acquires the force situation of the component to be installed and transmits it to the robot compliant control model. The binocular vision unit acquires the position of the visual target points and acquires the motion state of the component based on the position of the visual target points, and sends it to the robot compliant control model. The robot compliant control model processes the acquired force situation, acquires the correction information of the industrial robot displacement, sends it to the industrial robot, and acquires the robot's corrected Cartesian space pose information. The Cartesian space pose information is combined with the component motion state acquired by the binocular vision unit to obtain the stability of the robot system. The robot compliant control model is adjusted according to the stability to achieve the vibration suppression function of the system.

[0040] By adjusting the sliding mean filter window size based on the perception of the system's stable operation and continuous changes, the above settings ensure a rapid response to external force information during stable system operation, while avoiding rapid changes in force and displacement correction information caused by continuous and rapid changes in the system's state, thus achieving vibration suppression through force-position coupling control.

[0041] In a further embodiment, the robot compliance control model includes an adaptive moving average filtering unit, a force-position coupling control model, and a moving average filtering unit. The adaptive moving average filtering unit receives information from a six-dimensional force sensor and performs moving average filtering on multiple sets of sensor data, using the following formula: ,in This is the nth set of force data input into the force-potential coupling control model after filtering. To adapt the window size of the moving average filter, F represents the force sensor data before filtering. The force data obtained after moving average filtering is input into the force-position coupling control model. The force-position coupling control model calculates and outputs displacement correction information for the robot in six directions based on the obtained force data. This displacement correction information is then input into the moving average filtering unit. In a specific embodiment, the force-position coupling control model uses a model from the prior art, which will not be elaborated further here. The moving average filtering unit performs moving average filtering on the obtained displacement correction information, using the following formula: In the formula, Let D be the nth set of filtered displacement correction information, and let D be the displacement correction information output by the force-position coupling control. The filtered displacement correction information is then sent to the industrial robot. Through this setup, system status monitoring based on multi-source information, including force sensor information and visual tracking information, is introduced. This improves the accuracy of system status perception, avoids the instability or errors that may occur with single-source sensing information, and further ensures the vibration suppression effect of the force-position coupling control.

[0042] In another embodiment, the window of the adaptive moving mean filter The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, Through the above settings, a two-layer adaptive sliding mean filter is adopted, which combines information acquired by the force sensor with displacement correction information from the force-position coupling control output. This reduces the rapid changes in information from both the input and output perspectives, thereby reducing the possibility of vibration during the robot's force-position coupling control process.

[0043] In one further embodiment, force sensor data input into the force-position coupling control model is used for gravity compensation, zero drift compensation, and temperature drift compensation.

[0044] In a further embodiment, the motion state of the component acquired by the binocular vision unit is subjected to sliding mean filtering, as shown in the formula: In the formula, This represents the filtered nth group of robot end-effector pose information acquired through binocular vision. denoted as the sampling frequency of the binocular vision system, M represents the robot end-effector pose information acquired by the binocular vision system before filtering, and NM represents the size of the adaptive sliding mean filter window for the visual information.

[0045] In another embodiment, the basis for determining the stability of the system is:

[0046] If the robot's pose information in Cartesian space changes continuously and stably, and the pose information of the end effector monitored by binocular vision changes continuously and stably, then the system will operate stably.

[0047] If the robot's pose information in Cartesian space changes continuously and stably, while the pose information of the end effector changes rapidly and repeatedly, the system will operate stably.

[0048] If the robot's pose information in Cartesian space changes rapidly and repeatedly, while the pose information of the binocular vision recognition and monitoring end effector changes continuously and stably, then the system is unstable.

[0049] If the robot's Cartesian spatial pose information changes rapidly and repeatedly, and the binocular vision recognition monitoring end-effector pose information also changes rapidly and repeatedly, the system will be unstable. In one embodiment, rapidly and repeatedly changing end-effector pose information refers to rapid shaking of the robotic arm's end effector.

[0050] According to the second aspect, a method for vibration suppression in compliant assembly using industrial robots in human-robot collaboration is provided, comprising the following steps:

[0051] Step 1: Obtain the force information of the workpiece to be assembled;

[0052] Step two: Apply adaptive moving average filtering to the force information;

[0053] The filtered force information is used to calculate the displacement correction information of the industrial robot;

[0054] The displacement correction information, after being filtered by the moving average, is sent to the industrial robot to control its movement.

[0055] Step 3: Obtain the Cartesian space pose information of the industrial robot;

[0056] Obtain the end-effector pose information of the industrial robot based on binocular vision;

[0057] Step 4: Based on the Cartesian space pose information and end-effector pose information obtained in Step 1, determine whether the industrial robot system is operating stably. If yes, proceed to Step 1; otherwise, proceed to Step 2 to correct the filtering parameters.

[0058] In a further embodiment, in step two, the formula for adaptive moving mean filtering is: ,in This is the nth set of force data after filtering. The window size for adaptive moving average filtering is denoted by F, where F represents the force sensor data before filtering.

[0059] In another embodiment, the window of the adaptive moving mean filter The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, .

[0060] To gain a better understanding of the industrial robot human-machine collaborative compliant assembly vibration damping device and method provided by the present invention, a detailed description is provided below with reference to specific examples and accompanying drawings.

[0061] The specific implementation plan uses an ATI six-dimensional force sensor, a KUKA six-joint robot, a C-Track binocular vision system, and large cylindrical components. The specific steps are as follows:

[0062] Step 1: Determine the compliant assembly control logic for industrial robot human-robot collaboration: External forces are sensed by a six-dimensional force sensor and input into the robot's compliant control model. The control model outputs robot pose adjustment amounts to achieve compliant assembly through human-robot collaboration. See the flowchart for the specific human-robot collaborative force-position coupling control. Figure 1 .

[0063] In the above steps, when the six-dimensional force sensor senses external forces, it is necessary to eliminate the influence of the end effector and the load of the large components being held. Therefore, the external forces input into the robot's compliant control model are gravity-compensated force information, which includes torques in three directions and forces in three directions.

[0064] Simultaneously, based on the correlation between the force sensor coordinate system and the robot flange coordinate system, a mapping relationship is established between the forces and torques collected by the force sensor in three directions and the translation and rotation amounts in three directions of the robot's Cartesian space. In the specific implementation scheme of this patent, the direction of the force sensor coordinate system is consistent with the direction of the robot flange coordinate system, see [link to patent]. Figure 2 Therefore, the forces in the three directions correspond one-to-one with the translations in the three directions, and the torques in the three directions correspond one-to-one with the rotations in the three directions.

[0065] Step 2: Since the force information from the ATI force sensor is the input source for the entire robot's compliant assembly control, it is necessary to ensure the stability of the input source information and avoid vibrations in the force-position coupling assembly process caused by sudden changes in force information. Therefore, the external force information collected by the force sensor is first filtered. By performing sliding mean filtering on multiple sets of force sensor data, the sudden changes in force information in the input force-position coupling control model are reduced. This patent uses adaptive sliding mean filtering for force information processing, as detailed below:

[0066] (1)

[0067] In the formula, This is the nth set of force data input to the controller after filtering. The window size for adaptive moving average filtering is denoted by F, where F represents the force sensor data before filtering.

[0068] Since the sampling frequency of the force sensor does not correspond one-to-one with the robot control cycle, the moving average filter window needs to be constrained to ensure that there is data input for each control cycle of the robot.

[0069] (2)

[0070] In the formula, To input force information and determine the sampling frequency of the force sensor, it is typically set to 5000Hz. To ensure compliant control of the KUKA robot, a control cycle of 4ms is set. Size of the sliding mean filter window for the displacement information output by the compliance control model. These are the window correction parameters based on motion state.

[0071] (3)

[0072] In addition, the force sensor data input into the force-position coupling control model in step 2 also requires gravity compensation, zero drift compensation, and temperature drift compensation to ensure that the input force information is the force applied by the operator.

[0073] Step 3: After the adaptive sliding mean filtering in Step 2, the force information enters the force-position coupling control model. The displacement correction information of the robot in six directions is calculated and output. The displacement correction information is sent to the robot at this time. The robot performs pose correction according to the displacement correction information to realize force-position coupling compliant control in the assembly process.

[0074] In step 3, the displacement correction information output changes according to the input force sensor information, further inducing real-time correction of the robot's pose and achieving compliant control based on human operating intentions. Due to the large load of large components, the inertial forces generated by continuous and rapid changes in the robot's pose cause rapid changes in the force information collected by the force sensor, thus causing robot vibration. To avoid continuous and rapid changes in robot pose, a sliding mean filter is applied to the displacement correction information output by the force-position coupling control model.

[0075] (4)

[0076] In the formula, represents the nth group of displacement correction information after filtering, and D represents the displacement correction information output by the force-position coupling control.

[0077] Step 4: To ensure the response speed of the force-position coupling control of the industrial robot and to avoid vibration caused by rapid changes in motion state, the system operating state is divided into stable operation and continuous change. The window size of the sliding mean filter for the force information and displacement correction information collected by the force sensor is adaptively controlled.

[0078] (5)

[0079] In the formula, stable system operation is defined as a state in which the changing trends of the information from 5 consecutive sets of force sensors are consistent; continuous system change is defined as a state in which the changing trends of the information from 5 consecutive sets of force sensors are inconsistent, and the force information from two adjacent sets of force sensors alternately increases and decreases, with the change exceeding 15N.

[0080] Using formula (5), the sliding mean filtering of force sensor information and the sliding mean filtering of displacement correction information can be adaptively adjusted according to the system operating status to adjust the size of the filtering window.

[0081] Step 5: Force sensors, as a single source for determining the robot's force-position coupling control status, are susceptible to damage, which can affect the system's operational status assessment. Therefore, this project, based on force sensor judgment, introduces a binocular vision system to deploy target points on the surface of large components (see...). Figure 2 The system performs online identification, tracking, fitting, and monitoring of target points in the force-position coupling control process of the robot, judges the operating status of the robot system, and adjusts the size of the sliding mean filter window for force sensor acquisition information and displacement correction information to ensure rapid perception and vibration suppression of unstable operating states of the system.

[0082] In step 5, it is essential to ensure that the target point on the robot's end effector remains within the field of view of the binocular vision system. The binocular vision system determines the robot's pose in space by recognizing the target point information. The visually perceived state of the robot's end effector includes translational and rotational information in three directions. Since vision systems also exhibit instability, a moving average filter is required to apply to the visual information.

[0083] (6)

[0084] In the formula, This represents the filtered nth group of robot end-effector pose information acquired through binocular vision. is the sampling frequency of the binocular vision system, and M is the robot end-effector pose information acquired by the binocular vision system before filtering.

[0085] Step 6: A rapid system state identification method that integrates binocular vision system and robot end-effector pose information. By collecting Cartesian space pose information during the robot's force-position coupling control process and fusing it with the end-effector pose information monitored by binocular vision, the system stability is determined.

[0086] If the robot's pose information in Cartesian space changes continuously and stably, and the pose information of the end effector monitored by binocular vision changes continuously and stably, then the system will operate stably.

[0087] If the robot's pose information in Cartesian space changes continuously and stably, while the pose information of the end effector changes rapidly and repeatedly, the system will operate stably.

[0088] If the robot's pose information in Cartesian space changes rapidly and repeatedly, while the pose information of the binocular vision recognition and monitoring end effector changes continuously and stably, then the system is unstable.

[0089] If the robot's pose information in Cartesian space changes rapidly and repeatedly, and the pose information of the binocular vision recognition monitoring end effector changes rapidly and repeatedly, the system will be unstable.

[0090] Continuous and stable change refers to the change in pose information between two adjacent groups of robots not exceeding 0.5 mm, and the change in pose information of the end effector recognized by vision between two adjacent groups not exceeding 0.5 mm. Conversely, when the change in pose information between two adjacent groups of robots and the pose information recognized by vision exceeds 0.5 mm, and occurs consecutively in 5 or more groups, it is considered rapid and repetitive change.

[0091] When the robot's pose information and binocular vision recognition monitoring information are combined and the robot is determined to be unstable, the damping parameter of the system force-position coupling control model is increased, and the sliding mean filter of the force sensor information and displacement correction information under the unstable operation state of the system is simultaneously activated to avoid causing system vibration.

[0092] When the robot's pose information and binocular vision recognition and monitoring information are fused together to determine that the robot is running stably, the sliding mean filter window size of the current force sensor information and displacement correction information is not changed.

[0093] Step 7: During the human-robot collaborative assembly process of large components, steps 2 through 6 should be performed synchronously throughout (see...). Figure 3 This enables rapid vibration suppression of unstable operating states in human-machine collaboration, thereby improving the stability of system operation.

[0094] In summary, the industrial robot human-robot collaborative compliant assembly vibration damping device and method provided by this invention has at least the following advantages compared to the prior art:

[0095] 1) This invention proposes an adaptive sliding mean filtering vibration suppression method based on the operating state of a robot system. By sensing the stable operation and continuously changing state of the system, the size of the sliding mean filtering window is adjusted. This ensures a rapid response to external force information during the stable operation of the system, while avoiding rapid changes in force and displacement correction information caused by continuous and rapid changes in the system's state. This achieves vibration suppression through force-position coupling control.

[0096] 2) This invention employs a two-layer adaptive sliding mean filter that combines force sensor-acquired information with force-position coupling control output displacement correction information. This reduces rapid changes in information from both the input and output perspectives, thereby reducing the likelihood of vibration during the robot's force-position coupling control process.

[0097] 3) This invention introduces system status monitoring based on multi-source information such as force sensor information and visual tracking information, which improves the accuracy of system status perception, avoids the instability or error that may exist in single perception information, and further ensures the vibration suppression effect of force-position coupling control of the system.

[0098] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation beyond the orientation of the device as described in the figures. For example, if the device in the figures were inverted, a device described as "above" or "on top of" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0099] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore should not be construed as limiting the scope of protection of this invention.

[0100] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A vibration suppression device for human-robot collaborative compliant assembly of an industrial machine, characterized by, The system includes an industrial robot, a six-dimensional force sensor, a binocular vision unit, and a robot compliant control model. The industrial robot model meets the requirements for installing the component. The six-dimensional force sensor is mounted on the industrial robot flange and connected to the component via a mechanical connector. Several visual target points are set on the component. The six-dimensional force sensor acquires the force on the component and transmits it to the robot compliant control model. The binocular vision unit acquires the position of the visual target points and, based on the position, obtains the motion state of the component, which is then sent to the robot compliant control model. The robot compliant control model processes the acquired force information, obtains correction information for the industrial robot's displacement, sends it to the industrial robot, and acquires the robot's corrected Cartesian space pose information. By combining the Cartesian space pose information with the component motion state acquired by the binocular vision unit, the stability of the robot system is obtained. Based on the stability, the robot compliant control model is adjusted to achieve vibration suppression of the system.

2. The industrial robot human-machine collaborative compliant assembly vibration damping device according to claim 1, characterized in that, The robot compliance control model comprises an adaptive sliding mean filter unit, a force-position coupling control model and a sliding mean filter unit, the adaptive sliding mean filter unit receives information of a six-dimensional force sensor, performs sliding mean filtering on multiple groups of sensor data, and the formula is: Wherein, is the n th group of force data filtered and input into the force-position coupling control model, is the window size of adaptive sliding mean filtering, F is the force sensor data before filtering, the force data obtained after sliding mean filtering is input into the force-position coupling control model, the force-position coupling control model calculates the displacement correction information of six directions of the robot according to the obtained force data, the displacement correction information is input into the sliding mean filter unit, the sliding mean filter unit performs sliding mean filtering on the obtained displacement correction information, and the formula is: , wherein, is the n th group of displacement correction information after filtering, D is the displacement correction information output by the force-position coupling control, and the displacement correction information after filtering is sent to the industrial robot.

3. The industrial robot human-machine collaborative compliant assembly vibration damping device according to claim 2, characterized in that, The window of the adaptive moving mean filter The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, .

4. The industrial robot human-machine collaborative compliant assembly vibration damping device according to claim 1, characterized in that, Force sensor data input into the force-position coupling control model is used for gravity compensation, zero drift compensation, and temperature drift compensation.

5. The industrial robot human-machine collaborative compliant assembly vibration damping device according to claim 3, characterized in that, The motion state of the component acquired by the binocular vision unit is filtered using a sliding mean filter, as shown in the formula: In the formula, This is the robot end-effector pose information acquired by binocular vision after filtering. denoted as the sampling frequency of the binocular vision system, M represents the robot end-effector pose information acquired by the binocular vision system before filtering, and NM represents the size of the adaptive sliding mean filter window for the visual information.

6. The industrial robot human-machine collaborative compliant assembly vibration damping device according to claim 1, characterized in that, The criteria for judging the stability of a system are: If the robot's pose information in Cartesian space changes continuously and stably, and the pose information of the end effector monitored by binocular vision changes continuously and stably, then the system will operate stably. If the robot's pose information in Cartesian space changes continuously and stably, while the pose information of the end effector changes rapidly and repeatedly, the system will operate stably. If the robot's pose information in Cartesian space changes rapidly and repeatedly, while the pose information of the binocular vision recognition and monitoring end effector changes continuously and stably, then the system is unstable. If the robot's pose information in Cartesian space changes rapidly and repeatedly, and the pose information of the binocular vision recognition monitoring end effector changes rapidly and repeatedly, the system will be unstable.

7. A vibration suppression method using the compliant assembly vibration suppression device for industrial robot human-machine collaboration as described in any one of claims 1-6, characterized in that, The method includes the following steps: Step 1: Obtain the force information of the workpiece to be assembled; Step two: Apply adaptive moving average filtering to the force information; The filtered force information is used to calculate the displacement correction information of the industrial robot; The displacement correction information, after being filtered by the moving average, is sent to the industrial robot to control its movement. Step 3: Obtain the Cartesian space pose information of the industrial robot; Obtain the end-effector pose information of the industrial robot based on binocular vision; Step 4: Based on the Cartesian space pose information and end-effector pose information obtained in Step 1, determine whether the industrial robot system is operating stably. If yes, proceed to Step 1; otherwise, proceed to Step 2 to correct the filtering parameters.

8. The vibration suppression method according to claim 7, characterized in that, In step two, the formula for adaptive moving mean filtering is: ,in This is the nth set of force data after filtering. The window size for adaptive moving average filtering is denoted by F, where F represents the force sensor data before filtering.

9. The vibration suppression method according to claim 7, characterized in that, The window of the adaptive moving mean filter The constraint formula is: In the formula, To input force information and determine the sampling frequency of the force sensor, For robot compliant control cycle, Size of the sliding mean filter window for the displacement information output by the compliance control model. , For window correction parameters based on motion state, .

Images