Dual-axis hole compliant disassembly method and system for industrial robots

Abstract

The application discloses a kind of industrial robot double-shaft hole compliant disassembly method and system, wherein the method includes the following steps: three mechanism models of industrial robot double-shaft hole disassembly based on mathematical description are constructed;Industrial robot double-shaft hole compliant disassembly strategy model and its online training environment based on proximal policy optimization are constructed;According to the dynamic data in the online training environment, the state, action, reward item and penalty item suitable for the compliant disassembly problem of industrial robot double-shaft hole are set, and the compliant disassembly strategy model of industrial robot double-shaft hole is trained, wherein one of the penalty items is calculated based on the value of the three mechanism models in the disassembly process;According to the compliant disassembly strategy model of industrial robot double-shaft hole, the compliant disassembly of double-shaft hole is completed.The application can optimize the disassembly process of double-shaft hole, improve the applicability to batch double-shaft hole disassembly process, and effectively improve the generalization ability of the disassembly strategy of industrial robot double-shaft hole.

Description

Technical Field

[0001] This invention relates to the interdisciplinary research field combining intelligent manufacturing and artificial intelligence, and in particular to a compliant disassembly method and system for dual-axis holes of industrial robots based on near-end strategy optimization under dynamic parameters. Background Technology

[0002] In recent years, with the continuous acceleration of my country's industrialization and the increasing speed of product updates, a large amount of waste products are generated annually. If these massive quantities of waste products are not properly disposed of, it will not only cause enormous resource waste but also lead to serious environmental pollution. Recycling waste products and fully utilizing their residual value can significantly reduce environmental pollution. Furthermore, fully utilizing the value inherent in waste products and reinvesting it in the manufacturing process of new products can greatly improve resource utilization and effectively reduce the manufacturing costs of new products, which is of great significance for achieving the sustainable development of my country's manufacturing industry.

[0003] Dismantling is the primary step in recycling waste products, and industrial robot dismantling is an important means of promoting the intelligent development of waste product recycling. Depending on the type of parts, the dismantling process includes gear shaft dismantling, single-axis hole dismantling, and double-axis hole dismantling. Double-axis hole dismantling is a common step in the waste product dismantling process. When using industrial robots to dismantle double-axis holes, improper setting of the robot's motion control parameters can lead to excessive dismantling force / torque, often resulting in damage to the robot's end effector or the double-axis hole itself. This can cause damage to the industrial robot or render the double-axis hole unrecoverable.

[0004] Therefore, effectively reducing the disassembly force / torque during the disassembly of dual-axis holes by industrial robots and achieving compliant disassembly of dual-axis holes by industrial robots plays an important role in promoting the intelligence of the disassembly process of waste products and improving economic efficiency.

[0005] Due to variations in user habits and component maintenance strategies during product use, the physical parameters (dual-axis hole spacing, dual-axis hole radius, tilt angle, etc.) of batches of discarded products exhibit dynamic characteristics during disassembly. Furthermore, these physical parameters significantly influence disassembly force / torque. Ignoring the dynamic nature of these parameters and applying a compliant disassembly strategy based on static parameters to achieve compliant dual-axis hole disassembly under dynamic parameters may render the intended solution unsuitable. Therefore, proposing a compliant dual-axis hole disassembly strategy for industrial robots with dynamic parameters, while achieving batch disassembly of dual-axis holes, can improve the applicability of the disassembly process and effectively enhance the generalization ability of industrial robot dual-axis hole disassembly strategies. Summary of the Invention

[0006] The main objective of this invention is to provide a compliant disassembly method and system for dual-axis holes in industrial robots based on near-end strategy optimization under dynamic parameters, which can effectively improve the generalization ability of the disassembly strategy for dual-axis holes in industrial robots.

[0007] The technical solution adopted in this invention is:

[0008] A compliant disassembly method for a dual-axis bore in an industrial robot based on near-end strategy optimization under dynamic parameters is provided, comprising the following steps:

[0009] Three mechanistic models for the disassembly of a dual-axis hole in an industrial robot, based on mathematical description, are constructed. The first mechanistic model describes the distance traveled by a single-axis throughout the entire process of two-point contact within a single-axis hole. The second mechanistic model describes the distance traveled by both axes throughout the entire process of two-point contact within the outer side of the dual holes. The third mechanistic model describes the distance traveled by both axes throughout the entire process of two-point contact within the inner side of the dual holes.

[0010] A compliant disassembly strategy model for a dual-axis hole of an industrial robot based on near-end strategy optimization and its online training environment are constructed. The online training environment construction process is as follows: a 3D model of the dual axis and dual hole is constructed in the simulation environment, the industrial robot model is deployed in the simulation environment, and the dual-axis model is fixed on the force / torque sensor at the end of the industrial robot. By specifying the target position and attitude, the movement of the target point is controlled during the process of the industrial robot disassembling the dual axis from the dual hole, and the dynamic parameters during the disassembly process are obtained. The values ​​of three mechanism models during the disassembly process are calculated based on the dynamic parameters.

[0011] Based on dynamic data settings in the online training environment, states, actions, reward terms, and penalty terms are set for the compliant disassembly problem of a two-axis hole in an industrial robot. The compliant disassembly strategy model for a two-axis hole in an industrial robot is trained. One of the penalty terms is calculated based on the values ​​of three mechanism models during the disassembly process.

[0012] The compliant disassembly of the dual-axis hole was completed based on the compliant disassembly strategy model for industrial robots.

[0013] Following the above technical solution, the construction process of the three mechanism models is as follows: the mathematical relationship between each structural element is described based on the force balance relationship between the two axes and the two holes during the disassembly of the two-axis hole by the industrial robot under quasi-static conditions, and then the three mechanism models are constructed based on the mathematical relationship.

[0014] Following the above technical solution, the construction process of the dual-axis and dual-hole 3D model is as follows: create a cube model and a cylinder model in the simulation environment, and construct the dual-axis and dual-hole 3D model by combining the cylinder and the cube.

[0015] Following the above technical solution, in the process of building the online training environment, the inverse kinematics model of the UR5 robot is specifically constructed, and the movement of the industrial robot during the disassembly process is controlled by specifying the target position and posture.

[0016] Following the above technical solution, the penalty terms in the online training process of the industrial robot dual-axis hole compliant disassembly strategy model also include penalty terms for force and torque, penalty terms for the relative position of the two axes holes in the x and y directions, and penalty terms for the relative deflection angles of the two axes holes in the x, y, and z directions.

[0017] Following the above technical solution, the industrial robot dual-axis hole compliant disassembly strategy model continuously updates its parameters and continues training with the updated parameters.

[0018] Following the above technical solution, during the training process, the effectiveness of the current decomposition strategy model is evaluated by estimating the advantage function, and the decomposition strategy model is guided to be optimized and improved. In the optimization and improvement process, a pruning mechanism is introduced to limit the magnitude of policy updates, so as to ensure the smoothness of policy changes.

[0019] Following the above technical solution, while fixing the dual-axis model to the force / torque sensor at the end of the industrial robot, a virtual target point is generated and fixed on the central axis of the dual-axis.

[0020] This invention also provides a compliant disassembly system for a dual-axis hole in an industrial robot based on near-end strategy optimization under dynamic parameters, comprising:

[0021] The mechanism model construction module is used to construct three mechanism models for the disassembly of a dual-axis hole in an industrial robot based on mathematical description. The first mechanism model describes the distance moved by the single-sided axis throughout the entire process of maintaining two-point contact within the single-sided hole; the second mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the outer side within the dual holes; and the third mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the inner side within the dual holes.

[0022] The disassembly strategy model construction and training module is used to build a compliant disassembly strategy model for a dual-axis hole of an industrial robot based on near-end strategy optimization and its online training environment. The online training environment construction process is as follows: a 3D model of the dual axis and dual hole is built in the simulation environment, the industrial robot model is deployed in the simulation environment, and the dual-axis model is fixed on the force / torque sensor at the end of the industrial robot; by specifying the target position and attitude, the movement of the target point is controlled during the process of the industrial robot disassembling the dual axis from the dual hole, and the dynamic parameters during the disassembly process are obtained. The values ​​of three mechanism models during the disassembly process are calculated based on the dynamic parameters.

[0023] The compliant disassembly module is used to complete the compliant disassembly of dual-axis holes based on the compliant disassembly strategy model for industrial robots.

[0024] The present invention also provides a computer storage medium storing a computer program executable by a processor, the computer program executing the above-described method for compliant disassembly of a dual-axis hole in an industrial robot based on proximal strategy optimization under dynamic parameters.

[0025] The beneficial effects of this invention are as follows: The industrial robot dual-axis hole compliant disassembly strategy model constructed by this invention is based on the Proximal Policy Optimization (PPO) algorithm, which solves the continuous motion space control problem. Furthermore, this invention pre-constructs mechanism models for three scenarios in which dual-axis holes easily cause excessive disassembly force in the industrial robot during the disassembly process, and calculates one of the penalty terms for training the disassembly strategy model by calculating the mechanism model value in the online training environment, so as to avoid the three scenarios of excessive disassembly force as much as possible and optimize the disassembly process of dual-axis holes.

[0026] Furthermore, to address the issue of low compliance in the dual-axis hole disassembly process of industrial robots, a data interaction method between the dual-axis hole disassembly simulation environment and the deep reinforcement learning algorithm is combined. During online training, force and torque penalty terms are used to reduce the disassembly force / torque in the dual-axis hole disassembly process, effectively improving the compliance of the industrial robot's dual-axis hole disassembly process.

[0027] Furthermore, in the process of building the online training environment, specifically by constructing the inverse kinematics model of the UR5 robot, the movement of the industrial robot during the disassembly process is controlled by specifying the target position and posture, thereby generating biaxial hole disassembly data for training.

[0028] Furthermore, during training, the effectiveness of the current decomposition strategy model is evaluated by estimating the advantage function, and the decomposition strategy model is guided to be optimized and improved. In the optimization and improvement process, a pruning mechanism is introduced to effectively limit the magnitude of policy updates and ensure the smoothness of policy changes.

[0029] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1AThis is a flowchart of a compliant disassembly method for a dual-axis hole of an industrial robot based on near-end strategy optimization under dynamic parameters according to an embodiment of the present invention.

[0032] Figure 1B This is a flowchart of a compliant disassembly method for a dual-axis hole of an industrial robot based on near-end strategy optimization under dynamic parameters, according to another embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram of a dual-axis hole compliant disassembly model according to an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of four contact states during the disassembly of the biaxial hole of the present invention;

[0035] Figure 4 This is a schematic diagram of the initial state of disassembly of the dual-axis hole of the present invention;

[0036] Figure 5 This is a schematic diagram of the two-point contact state during the disassembly of the dual-axis hole of the present invention;

[0037] Figure 6 This is a schematic diagram of two-point contact of a single-axis hole in this invention;

[0038] Figure 7 This is a schematic diagram of two-point contact on the outer side of the present invention;

[0039] Figure 8 This is a schematic diagram of two points of contact on the inner side of the present invention;

[0040] Figure 9 This is a schematic diagram of the structural framework of the industrial robot dual-axis hole compliant disassembly strategy based on near-end strategy optimization according to the present invention;

[0041] Figure 10 This is a schematic diagram showing the relationship between the biaxial hole disassembly force / torque and depth under dynamic parameters in test group 1 of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0043] It should be noted that the illustrations provided in the embodiments of the present invention are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0044] In this invention, it should also be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first" and "second" are used only for descriptive and distinguishing purposes and should not be construed as indicating or implying relative importance.

[0045] Furthermore, it should be noted that the features of the various embodiments of the present invention can be combined or integrated in whole or in part, and as those skilled in the art will understand, they can interact and operate in different ways. Each embodiment can be implemented independently of each other or in association with one another.

[0046] like Figure 1A As shown, the compliant disassembly method for dual-axis holes in industrial robots based on near-end strategy optimization under dynamic parameters according to an embodiment of the present invention includes the following steps:

[0047] S1. Construct three mechanistic models for the disassembly of a dual-axis hole in an industrial robot based on mathematical description. The first mechanistic model describes the distance moved by the single-axis throughout the entire process of two-point contact within the single-axis hole. The second mechanistic model describes the distance moved by the dual axes throughout the entire process of two-point contact within the dual holes. The third mechanistic model describes the distance moved by the dual axes throughout the entire process of two-point contact within the dual holes.

[0048] S2. Construct a compliant disassembly strategy model for a dual-axis hole of an industrial robot based on near-end strategy optimization and its online training environment. The online training environment is constructed as follows: a 3D model of the dual axis and dual hole is built in the simulation environment, the industrial robot model is deployed in the simulation environment, and the dual-axis model is fixed on the force / torque sensor at the end of the industrial robot. By specifying the target position and attitude, the movement of the target point is controlled during the process of the industrial robot disassembling the dual axis from the dual hole, and the dynamic parameters during the disassembly process are obtained. The values ​​of the three mechanism models during the disassembly process are calculated based on the dynamic parameters.

[0049] S3. Based on the dynamic data in the online training environment, set the state, action, reward and penalty terms suitable for the dual-axis hole compliant disassembly problem of industrial robots, and train the dual-axis hole compliant disassembly strategy model of industrial robots. One of the penalty terms is calculated based on the values ​​of the three mechanism models during the disassembly process.

[0050] S4. Complete the compliant disassembly of the dual-axis hole based on the compliant disassembly strategy model for industrial robots.

[0051] This invention addresses the challenge of smoothly disassembling numerous batches of biaxial holes using industrial robots due to the dynamic nature of parameters such as the surface friction coefficient, hole spacing, shaft radius, and hole radius. The proposed strategy, based on proximal-end policy optimization under dynamic parameters, effectively improves the efficiency and generalization of biaxial hole disassembly, promoting the intelligent development of waste product disassembly. Considering the dynamic nature of these parameters, this invention constructs a mathematically described mechanism model for industrial robot biaxial hole disassembly, analyzing the two-point contact range under different parameters to provide a theoretical basis for the smooth disassembly strategy. Based on this, and combined with a biaxial hole disassembly simulation environment, a deep reinforcement learning algorithm based on proximal-end policy optimization is used to construct a smooth disassembly strategy model for industrial robots. This invention provides theoretical and technical support for advancing the intelligentization of waste product disassembly processes.

[0052] The Proximal Policy Optimization (PPO) algorithm described in the above embodiments is a typical deep reinforcement learning algorithm, mainly used to solve continuous motion space control problems. This invention applies it to the problem of compliant disassembly strategy for dual-axis holes in industrial robots under dynamic parameters. The compliant disassembly strategy for dual-axis holes in industrial robots can dynamically adjust the robot's motion control parameters based on the disassembly force / torque sensed by force sensors and the robot's end-effector pose during the disassembly process, minimizing the disassembly force / torque. By combining the state data fed back during the disassembly process, the agent gradually learns how to compliantly disassemble dual-axis holes, thereby constructing a compliant disassembly strategy for dual-axis holes in industrial robots under dynamic parameters.

[0053] Furthermore, this invention pre-constructs a mechanism model for three scenarios where the disassembly force of the industrial robot is too large during the disassembly process of the biaxial hole, and influences one of the penalty terms of the training disassembly strategy model by calculating the mechanism model value in the online training environment, so as to avoid the three scenarios of excessive disassembly force as much as possible and optimize the disassembly process of the biaxial hole.

[0054] Furthermore, the construction process of the three mechanism models is as follows: the mathematical relationship between each structural element is described based on the force balance relationship between the two axes and the two holes during the disassembly of the two-axis hole by the industrial robot under quasi-static conditions, and then the three mechanism models are constructed based on the mathematical relationship.

[0055] Furthermore, the construction process of the biaxial and bi-hole 3D model is as follows: a cube model and a cylinder model are created in the simulation environment, and the biaxial and bi-hole 3D model is constructed by combining the cylinder and the cube.

[0056] Furthermore, in the process of building the online training environment, specifically by constructing the inverse kinematics model of the UR5 robot, the movement of the industrial robot during the disassembly process is controlled by specifying the target position and posture.

[0057] Furthermore, the penalty terms in the online training process of the industrial robot dual-axis hole compliant disassembly strategy model also include penalty terms for force and torque, penalty terms for the relative position between the two axes in the x and y directions, and penalty terms for the relative deflection angle between the two axes in the x, y, and z directions.

[0058] Furthermore, the industrial robot dual-axis hole compliant disassembly strategy model is continuously updated with parameters, and the updated parameters are used for further training.

[0059] Furthermore, during training, the effectiveness of the current decomposition strategy model is evaluated by estimating the advantage function, and the decomposition strategy model is guided to be optimized and improved. In the optimization and improvement process, a pruning mechanism is introduced to limit the magnitude of policy updates, so as to ensure the smoothness of policy changes.

[0060] Furthermore, while fixing the dual-axis model to the force / torque sensor at the end of the industrial robot, a virtual target point is generated and fixed on the dual-axis centerline.

[0061] In another embodiment of the present invention, a first mechanism model Lh1 for the two-point contact case of a single-axis hole, a second mechanism model Lh2 for the two-point contact case on the outer side, and a third mechanism model Lh3 for the two-point contact case on the inner side are constructed based on the mathematical relationships between the various structural elements in the dual-axis hole. Addressing the issue of low compliance in the disassembly process of dual-axis holes in industrial robots, a data interaction method between the dual-axis hole disassembly simulation environment and the deep reinforcement learning algorithm is combined. A deep reinforcement learning algorithm based on proximal policy optimization is used to train and generate a compliant disassembly strategy model for dual-axis holes, reducing the disassembly force / torque in the disassembly process and effectively improving the compliance of the industrial robot's dual-axis hole disassembly process.

[0062] like Figure 1B As shown in the figure, the compliant disassembly method for dual-axis holes in industrial robots based on near-end strategy optimization under dynamic parameters proposed in this embodiment mainly includes the following steps:

[0063] (1) Construct a mathematically described model of the disassembly mechanism of a dual-axis hole in an industrial robot. Based on the force balance relationship between the dual axes and the dual holes during the disassembly process under quasi-static conditions, describe the distance L from the compliant center of the industrial robot to the center of the bottom of the dual axes. CThe model includes the shaft hole radius R / r, friction coefficient μ, and hole spacing Ld. Based on the mathematical relationship between these multiple factors, a mathematically described industrial robot dual-axis hole disassembly mechanism model is constructed, including the first mechanism model Lh1 under the single-axis hole two-point contact case, the second mechanism model Lh2 under the outer two-point contact case, and the third mechanism model Lh3 under the inner two-point contact case.

[0064] (2) A compliant disassembly strategy model for a dual-axis hole in an industrial robot based on a deep reinforcement learning algorithm with proximal policy optimization is proposed. Considering the dynamic nature of the parameters in the dual-axis hole disassembly process of the industrial robot, a deep reinforcement learning algorithm environment based on proximal policy optimization for the compliant disassembly strategy of a dual-axis hole in an industrial robot is constructed using the CoppeliaSim simulation environment. State, action, and reward functions suitable for the compliant disassembly problem of a dual-axis hole in an industrial robot are set, and a compliant disassembly strategy model for a dual-axis hole in an industrial robot is generated through training. Subsequently, the dual-axis hole parameters are dynamically adjusted, and the operating effect of the compliant disassembly strategy model for a dual-axis hole in an industrial robot under dynamic parameters is analyzed.

[0065] In step (1), the distance L from the compliant center of the industrial robot to the center of the bottom of the dual axes is constructed. C The mathematical relationships between the shaft hole radius R / r, friction coefficient μ, hole spacing Ld, and the contact distances Lh1, Lh2 (outer side), and Lh3 (inner side) of a single shaft hole are used to construct a disassembly mechanism model for a dual-axis hole in an industrial robot. The specific process is as follows:

[0066] Industrial robot dual-axis hole compliant disassembly mechanism model as follows Figure 2 As shown, L C It is the distance from the center of compliance to the center of the bottom of the dual shaft, R / r are the hole radius and shaft radius respectively, Ld is the hole spacing, and O C K represents the center of compliance. x and K θ These represent lateral stiffness and angular stiffness, respectively. During the disassembly of a dual-axis hole in an industrial robot, there may be several contact states between the dual axes and the dual holes, specifically no contact, line contact, one-point contact, and two-point contact, etc. Figure 3 As shown in the diagram. When the dual-axis hole is in a two-point contact state, it can easily cause excessive disassembly force by the industrial robot. The initial state of the industrial robot disassembling the dual-axis hole is as follows: Figure 4 As shown, the initial tilt angle of the dual shaft is θ0, the initial horizontal distances from the compliant center and the bottom center of the shaft to the central axis of the hole are U0 and ε0, respectively, the initial disassembly depth of the dual shaft hole is h0, the lateral error is δ0, and the angular error is β0.

[0067] When two points of contact occur during the disassembly of a dual-axis hole in an industrial robot, they can be further categorized into three types: single-axis hole two-point contact, outer two-point contact, and inner two-point contact. Figure 5 As shown.

[0068] ① Two-point contact of a single-axis hole

[0069] The two-point contact state of a single-axis hole during the disassembly process of an industrial robot with a dual-axis hole, such as... Figure 6 As shown. During the disassembly of the dual-axis hole, the physical quantities such as the horizontal distance U from the compliance center to the hole's central axis, the horizontal distance ε from the center of the shaft's bottom end to the hole's central axis, and the shaft's tilt angle θ satisfy the following mathematical relationship:

[0070] U+ε=L C sinθ (1)

[0071] U-U0+(ε-ε0)=L C (θ-θ0) (2)

[0072]

[0073] U-U0=L C (θ-θ0) (4)

[0074] When the two points of a single-axis hole are in contact, the geometric constraint mathematical model of the biaxial hole at a small angle θ can be described by formula (5):

[0075]

[0076] When the contact changes from two points to one point, combining the force and torque generated by the compliant center, as well as the force and torque conditions of the biaxial shaft during disassembly, F can be derived. x Description of the M parameter:

[0077] F x =-F N1 =-Kx(δ0+U0-U) (6)

[0078] M=(h-μr)F N1 =KxL C (δ0+U0-U)+K θ (β0+θ-θ0)(7)

[0079] Combining the above formulas (1) to (7), we get

[0080] αh 2 +βh+γ=0(8)

[0081] α=Kxδ0+KxL C θ0 (9)

[0082] β=(θ0-β0)(K θ -KxL C 2 -μrKxL C)-(L C +μr)Kx(L C β0+δ0)-2(Rr)KxL C (10)

[0083] γ=-2cR(K θ -KxL C 2 -μrKxL C (11)

[0084] The critical depth h of two-point contact in a single-axis hole is one of the two solutions to equation (8), and the contact distance Lh1 between the two points can be described as:

[0085]

[0086] ② Two points of contact on the outer side

[0087] Two-point contact on the outer side of a biaxial hole can be equivalently transformed into two-point contact in a uniaxial hole, such as... Figure 7 As shown, the following relationship is satisfied:

[0088]

[0089] During the disassembly of a dual-axis hole, the coupling relationship of the compliant center motion can be described by the following formula:

[0090] U+ε= L C sinθ (14)

[0091]

[0092] When the two points are in contact on the outside, the geometric constraint mathematical model of the biaxial hole at a small angle θ can be described by formula (17):

[0093]

[0094] When the contact changes from two points to one point, combining the force and torque generated by the compliant center, as well as the force and torque conditions of the biaxial shaft during disassembly, F can be derived. x Description of the M parameter:

[0095] F x =-F N1 =-Kx(δ0+U0-U) (18)

[0096] M=(h-μr′)F N1 =Kx L C (δ0+U0-U)+K θ (β0+θ-θ0)(19)

[0097] Combining the above formulas (13) to (19), we get

[0098] α′h 2 +β′h+γ′=0(20)

[0099] α′=Kx(δ0+c′R′-ε0)+KxL C θ0 (21)

[0100] β′=(θ0-β0)(K θ -KxL C 2 -μr′KxL C )-(L C +μr′)Kx(δ0+2c′R′-ε0) (22)

[0101] γ′=-2c′R′(K θ -KxL C 2 -μr′KxL C ) (twenty three)

[0102] The critical depth h of the two outer points in contact is given by two solutions to equation (20), and the contact distance Lh2 between the two points in this state is:

[0103]

[0104] ③ Two points of contact on the inner side

[0105] The two-point contact on the inner side during the disassembly process of a dual-axis hole in an industrial robot can be considered as an inverted single-axis hole. For example... Figure 8 As shown, the shaded area represents an equivalent axis that satisfies the following relationship:

[0106]

[0107] The motion coupling relationship of the compliant center of the industrial robot during the disassembly of a dual-axis hole is described by the following formula:

[0108] U+ε=L C sinθ (26)

[0109] ε=r″cosθ-R″ (27)

[0110] U-U0=L C (θ-θ0) (28)

[0111] When the two points are in contact on the inner side, the geometric constraint mathematical model of the biaxial hole at a small angle θ can be described by formula (31):

[0112]

[0113] When the contact changes from two points to one point, combining the force and torque generated by the compliant center, as well as the force and torque conditions of the biaxial shaft during disassembly, F can be derived. x Description of the M parameter:

[0114] F x =-F N2 =-Kx(δ0+U0-U) (30)

[0115] M = μR″F N2 =Kx(L C -h)(δ0+U0-U)+K θ (β0+θ-θ0)(31)

[0116] Combining the above formulas (25) to (31), we get:

[0117] α″h 2 +β″h+γ″=0(32)

[0118] α″=Kx(δ0+L C θ0) (33)

[0119] β″=(θ0-β0)(K θ -KxL C 2 -μR″KxL C )-(L C +μR″)Kx(δ0+L C β0)+2c″R″KxL C (34)

[0120] γ″=-2c′R′(K θ -KxL C 2 -μr′KxL C (35)

[0121] The critical depth of contact between the two inner points during the disassembly process of the dual-axis hole of the industrial robot is given by two solutions of formula (32), and the contact distance Lh3 between the two points is:

[0122]

[0123] In step (2), to achieve data interaction between the industrial robot dual-axis hole compliant disassembly simulation environment and the deep reinforcement learning algorithm, an online training environment for the deep reinforcement learning algorithm for multiple types of dual-axis hole compliant disassembly processes of industrial robots is constructed. The running effect of the industrial robot dual-axis hole compliant disassembly strategy model under dynamic parameters is analyzed. The specific operations are as follows:

[0124] ① Implement data interaction between PyCharm software and the CoppeliaSim simulation environment. First, create a cube model and a cylinder model in the CoppeliaSim simulation environment. Construct a dual-axis, dual-hole 3D model by combining the cylinder and cube. Then, deploy the UR5 robot model to the CoppeliaSim simulation environment. Fix the dual-axis model to the force / torque sensor at the UR5 robot's end effector. Simultaneously, generate a dummy point UR5_target and fix it on the dual-axis central axis, making it a child of the dual-axis. Construct the inverse kinematics model of the UR5 robot, enabling control of the industrial robot's disassembly process by specifying the target position and orientation of UR5_target. In the CoppeliaSim simulation environment, call the sim.simxStart and sim.simxStartSimulation functions to initiate data interaction between the CoppeliaSim server and the PyCharm client. The sim.simxSetObjectPosition and sim.simxSetObjectQuaternion functions can specify the target's spatial coordinates and quaternions to control the target's position and orientation, respectively. In PyCharm, calling the `sim.simxSetObjectPosition` and `sim.simxSetObjectQuaternion` functions enables data transfer from the PyCharm environment to the CoppeliaSim simulation environment for dual-axis hole disassembly operations, controlling the movement of target points during the industrial robot disassembly process. The `sim.simxGetObjectPosition`, `sim.simxGetObjectOrientation`, and `sim.simxReadForceSensor` functions can read the target's spatial coordinates, angle, and force sensor data, respectively, thereby enabling the data transfer of dual-axis hole disassembly data (such as disassembly force / torque, relative position of the dual-axis holes, and relative attitude of the dual-axis holes) from the CoppeliaSim simulation environment to the PyCharm environment.

[0125] ② Establish a compliant disassembly strategy model for dual-axis holes in industrial robots based on proximal strategy optimization. The PPO algorithm's elements include state (s), action (a), and reward (r); the intelligent system determines the appropriate action based on the current state s. t Choose the optimal disassembly action a t Among them, state s t It is a one-dimensional vector of length 12, as shown in formula (37). Where rp x ,rp y and rp zThese represent the relative positions of the biaxial holes in the x, y, and z directions, respectively. α ra β and ra γ These represent the relative deflection angles of the biaxial bore in the x, y, and z directions, respectively. x f y f z m x m y and m z These are the forces and torques in the x, y, and z directions read by the force / torque sensor, respectively. max O max F max and M max The values ​​are set to 3 mm, 0.03 rad, 100 N, and 3 N·m, respectively. The action is a one-dimensional vector of length 6, as shown in formula (38), po x ,po y ,po z 、ao x 、ao y and ao z These represent the position offset and angle offset of the robot's end effector (UR5_target), respectively. x ,po y The range is -0.1mm to 0.1mm, po z The range is -1.5mm to 1.5mm, ao x 、ao y and ao z The range is -0.005 rad to 0.005 rad. The reward is defined by formulas (39) to (46). Where, P F and P M P represents the penalty terms for force and torque, respectively. p P represents the penalty term for the relative positional quantity between the two axial holes in the x and y directions. a P represents the penalty term for the relative deflection angles in the x, y, and z directions between the two-axis holes. Lh R represents the penalty term for the contact distance between two points in a dual-axis hole. H The reward item, R, represents the distance of the dual-axis disassembly. S The reward item for successful dual-axis disassembly, Lh max It is set to 0.5m, h represents the disassembly depth, H represents the double-hole depth, and a n This represents the total number of actions in the current round. Lh1, Lh2, and Lh3 are obtained from formulas (9)~(12), (21)~(24), and (33)~(36), respectively, where θ0=0rad, δ0=2mm, β0=0rad, and K x =4N / mm, K θ= 30 Nmm / rad, L C R, r, μ, and Ld are determined by the biaxial hole parameters. If a is satisfied... n >100, P F ≤-1, P M ≤-1, P p ≤-1 or P a If h ≤ -1.2, it means the disassembly task of the current round has failed. If h ≥ H, it means the disassembly task of the current round has succeeded. Both require the termination condition information done = 1, the environment of the PPO algorithm is reset, and a new disassembly process starts again.

[0126]

[0127] action(a) = [po x ,po y ,po z ,ao x ,ao y ,ao z (38)

[0128] reward(r) = P F +P M +P p +P a +P Lh +R H +R S (39)

[0129]

[0130] The PPO algorithm collects experience samples through the interaction between the agent and the environment, and optimizes the strategy based on these samples. The structural framework of the compliant disassembly strategy for a dual-axis hole in an industrial robot based on proximal policy optimization is as follows: Figure 9 As shown. After the policy update is completed, the PPO algorithm discards the previously used experience samples and uses the updated policy to collect new sample data to ensure that the data used during training is consistent with the current policy. In the PPO algorithm, the policy network is responsible for generating the actions of the industrial robot's dual-axis hole disassembly process, and the evaluation network evaluates the effect of the current policy network by estimating the advantage function and guides the policy network to improve. The loss function of the PPO algorithm is shown in formula (47). Where ∈ is the clipping coefficient, π old The parameter is θ old strategy, This is the dominance function. By introducing this pruning mechanism, the magnitude of policy updates can be effectively limited, ensuring the smoothness of policy changes.

[0131]

[0132] The hyperparameters used in this patent during the training process are shown in Table 1.

[0133] Table 1 PPO Hyperparameters

[0134]

[0135] First, the policy network uses the current state s t Generate action strategy a t Through action strategy a t Control the industrial robot to perform disassembly movements, while the environment updates its status. t+1 And give the calculated reward value r. t , tuple t ,a t ,s t+1 ,r t The experience is stored in an experience pool. When the number of experiences in the experience pool reaches the number of samples in the batch, the experience is taken from the experience pool and used to update the neural network parameters. The updated policy network then continues to generate action policies.

[0136] ③ Implement a compliant disassembly strategy for dual-axis holes in industrial robots under dynamic parameters.

[0137] Table 2. Dual-axis bore parameters

[0138]

[0139] Based on the algorithm parameters in group ② of Table 2, a compliant disassembly strategy model for a dual-axis hole in an industrial robot was trained. Using the converged deep reinforcement learning model, compliant disassembly of the dual-axis hole under different parameters (①, ④, ⑦, ⑩ in Table 2) was achieved. The average values ​​m_F and m_T of the absolute values ​​of force and torque, as well as the median values ​​mid_F and mid_T of the absolute values ​​of force and torque, were measured during a single disassembly process. The converged strategy was used as test group 1, and the untrained strategy was used as test group 2. The results for the two test groups are shown in Table 3.

[0140] Table 3. Disassembly data of the dual-axis bore of the industrial robot under dynamic parameters.

[0141]

[0142] The results in Table 3 show that, under the same parameters, m_F, m_T, mid_F, and mid_T of test group 1 are lower than those of test group 2, indicating that the converged depth enhancement model helps reduce the disassembly force and torque during the biaxial hole disassembly process of the industrial robot. The relationship between the biaxial hole disassembly force / torque and depth in test group 1 under dynamic parameters is as follows: Figure 10 As shown, the strategy model exhibits good compliance during the biaxial hole disassembly process under dynamic parameters.

[0143] ​The industrial robot dual-axis hole disassembly mechanism model constructed in this embodiment provides a theoretical basis for proposing a compliant disassembly strategy for industrial robots with dual-axis holes under dynamic parameters. A deep reinforcement learning algorithm environment based on proximal policy optimization for the compliant disassembly strategy of industrial robots with dual-axis holes was constructed using the CoppeliaSim simulation environment. Combining elements such as state, action, and reward applicable to the dual-axis hole compliant disassembly problem, a training-generated industrial robot dual-axis hole compliant disassembly strategy model based on the PPO algorithm was proposed. The operational effect of the industrial robot dual-axis hole compliant disassembly strategy model under dynamic parameters was analyzed, proving the effectiveness of the proposed compliant disassembly strategy.

[0144] The present invention provides a compliant disassembly system for a dual-axis hole in an industrial robot based on near-end strategy optimization under dynamic parameters, comprising:

[0145] The mechanism model construction module is used to construct three mechanism models for the disassembly of a dual-axis hole in an industrial robot based on mathematical description. The first mechanism model describes the distance moved by the single-sided axis throughout the entire process of maintaining two-point contact within the single-sided hole; the second mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the outer side within the dual holes; and the third mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the inner side within the dual holes.

[0146] The disassembly strategy model construction and training module is used to build a compliant disassembly strategy model for a dual-axis hole of an industrial robot based on near-end strategy optimization and its online training environment. The online training environment construction process is as follows: a 3D model of the dual axis and dual hole is built in the simulation environment, the industrial robot model is deployed in the simulation environment, and the dual-axis model is fixed on the force / torque sensor at the end of the industrial robot; by specifying the target position and attitude, the movement of the target point is controlled during the process of the industrial robot disassembling the dual axis from the dual hole, and the dynamic parameters during the disassembly process are obtained. The values ​​of three mechanism models during the disassembly process are calculated based on the dynamic parameters.

[0147] The compliant disassembly module is used to complete the compliant disassembly of dual-axis holes based on the compliant disassembly strategy model for industrial robots.

[0148] Each module is mainly used to implement the various steps of the above method embodiments, which will not be described in detail here.

[0149] This application also provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, server, App application store, etc., which stores a computer program, and the program implements corresponding functions when executed by a processor. The computer-readable storage medium of this embodiment implements a method for compliant disassembly of a dual-axis hole in an industrial robot under dynamic parameters based on near-end strategy optimization, when executed by a processor.

[0150] In summary, addressing the urgent need for waste product recycling and the difficulty of existing methods in effectively constructing a compliant disassembly strategy for dual-axis holes in industrial robots under dynamic parameters, this invention proposes a compliant disassembly strategy for dual-axis holes in industrial robots based on near-end strategy optimization under dynamic parameters. A mathematical model is constructed to describe the physical quantities such as the friction coefficient and hole distance of the dual-axis holes during the disassembly process with respect to the contact distance between the two points of the dual-axis holes. The proposed method provides theoretical support for promoting the intelligent development of waste product recycling processes and can actively respond to the urgent needs of my country's sustainable and green development.

[0151] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0152] The order of the steps in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0153] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A compliant disassembly method for a dual-axis hole in an industrial robot based on near-end strategy optimization under dynamic parameters, characterized in that, Includes the following steps: Three mechanistic models for the disassembly of a dual-axis hole in an industrial robot, based on mathematical description, are constructed. The first mechanistic model describes the distance traveled by a single-axis throughout the entire process of two-point contact within a single-axis hole. The second mechanistic model describes the distance traveled by both axes throughout the entire process of two-point contact within the outer side of the dual holes. The third mechanistic model describes the distance traveled by both axes throughout the entire process of two-point contact within the inner side of the dual holes. A compliant disassembly strategy model for a dual-axis hole of an industrial robot based on near-end strategy optimization and its online training environment are constructed. The online training environment construction process is as follows: a 3D model of the dual axis and dual hole is constructed in the simulation environment, the industrial robot model is deployed in the simulation environment, and the dual-axis model is fixed on the force / torque sensor at the end of the industrial robot. By specifying the target position and attitude, the movement of the target point is controlled during the process of the industrial robot disassembling the dual axis from the dual hole, and the dynamic parameters during the disassembly process are obtained. The values ​​of three mechanism models during the disassembly process are calculated based on the dynamic parameters. Based on dynamic data settings in the online training environment, a compliant disassembly strategy model for a two-axis hole in an industrial robot is trained, specifying states, actions, rewards, and penalties applicable to this problem. States include the relative position, relative deflection angle, force, and torque of the two-axis holes; actions include the position and angular offset of the robot's end effector; penalties include penalties for force and torque, penalties for the relative position of the two-axis holes, penalties for the relative deflection angle of the two-axis holes, and penalties for the two-point contact distance of the two-axis holes; rewards include rewards for the two-axis disassembly distance and rewards for successful two-axis disassembly. The penalty for the two-point contact distance of the two-axis holes is calculated based on the values ​​of three mechanistic models during the disassembly process. The compliant disassembly of the dual-axis hole was completed based on the compliant disassembly strategy model for industrial robots.

2. The compliant disassembly method for dual-axis holes in industrial robots based on near-end strategy optimization under dynamic parameters as described in claim 1, characterized in that, The construction process of the three mechanism models is as follows: the mathematical relationship between each structural element is described based on the force balance relationship between the two axes and the two holes during the disassembly of the two-axis hole by the industrial robot under quasi-static conditions, and then the three mechanism models are constructed based on the mathematical relationship.

3. The compliant disassembly method for dual-axis holes in industrial robots based on proximal strategy optimization under dynamic parameters as described in claim 1, characterized in that, The construction process of the biaxial and bi-hole 3D model is as follows: create a cube model and a cylinder model in the simulation environment, and construct the biaxial and bi-hole 3D model by combining the cylinder and the cube.

4. The compliant disassembly method for dual-axis holes in industrial robots based on proximal strategy optimization under dynamic parameters as described in claim 1, characterized in that, In the process of building the online training environment, the inverse kinematics model of the UR5 robot is constructed, and the movement of the industrial robot during the disassembly process is controlled by specifying the target position and posture.

5. The compliant disassembly method for dual-axis holes in industrial robots based on proximal strategy optimization under dynamic parameters as described in claim 1, characterized in that, During the training process, the industrial robot's dual-axis hole compliant disassembly strategy model continuously updates its parameters and continues training with the updated parameters.

6. The compliant disassembly method for dual-axis holes in industrial robots based on proximal strategy optimization under dynamic parameters as described in claim 1, characterized in that, During training, the effectiveness of the current decomposition strategy model is evaluated by estimating the advantage function, and the decomposition strategy model is optimized and improved. In the optimization and improvement process, a pruning mechanism is introduced to limit the magnitude of policy updates, so as to ensure the smoothness of policy changes.

7. The compliant disassembly method for a dual-axis bore of an industrial robot based on proximal strategy optimization under dynamic parameters according to any one of claims 1-6, characterized in that, While fixing the dual-axis model to the force / torque sensor at the end of the industrial robot, a virtual target point is generated and fixed on the central axis of the dual-axis.

8. A compliant disassembly system for a dual-axis hole in an industrial robot based on proximal strategy optimization under dynamic parameters, characterized in that, include: The mechanism model construction module is used to construct three mechanism models for the disassembly of a dual-axis hole in an industrial robot based on mathematical description. The first mechanism model describes the distance moved by the single-sided axis throughout the entire process of maintaining two-point contact within the single-sided hole; the second mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the outer side within the dual holes; and the third mechanism model describes the distance moved by the dual axes throughout the entire process of maintaining two-point contact on the inner side within the dual holes. The disassembly strategy model construction and training module is used to build a compliant disassembly strategy model for a dual-axis hole of an industrial robot based on proximal strategy optimization and its online training environment. The online training environment construction process is as follows: A 3D model of the dual axes and dual holes is constructed in the simulation environment; the industrial robot model is deployed to the simulation environment; the dual-axis model is fixed to the force / torque sensor at the end effector of the industrial robot; by specifying the target position and attitude, the movement of the target point is controlled during the process of disassembling the dual axes from the dual holes, and dynamic parameters during the disassembly process are acquired. The values ​​of three mechanistic models during the disassembly process are calculated based on the dynamic parameters; and the dynamic data in the online training environment is used to... Set up states, actions, rewards, and penalties for the compliant disassembly of a two-axis hole in an industrial robot. Train a compliant disassembly strategy model for the two-axis hole. The states include the relative position of the two-axis holes, the relative deflection angle, force, and torque. The actions include the position and angle offset of the robot end effector. The penalties include penalties for force and torque, penalties for the relative position of the two-axis holes, penalties for the relative deflection angle of the two-axis holes, and penalties for the two-point contact distance of the two-axis holes. The rewards include a reward for the two-axis disassembly distance and a reward for successful two-axis disassembly. The penalty for the two-point contact distance of the two-axis holes is calculated based on the values ​​of three mechanism models during the disassembly process. The compliant disassembly module is used to complete the compliant disassembly of dual-axis holes based on the compliant disassembly strategy model for industrial robots.

9. A computer storage medium, characterized in that, It contains a computer program that can be executed by a processor, which performs the compliant disassembly method for a dual-axis hole of an industrial robot based on proximal strategy optimization under the dynamic parameters described in any one of claims 1-7.

Images