Joint elastomer optimization design method based on Gaussian filtering algorithm

Abstract

The invention relates to the technical field of robot joint elastomer optimization, in particular to a joint elastomer optimization design method based on a Gaussian filtering algorithm, and the method comprises the steps: building an Archimedes spiral-based finite element simulation model, constructing an optimization function with maximized deformation and minimized stress as targets, and optimizing the joint elastomer. Key design parameters of the elastomer are selected for iterative optimization in stages; and finally, an edge curve of an H-bridge structure of the elastomer is smoothed through a Gaussian filtering algorithm. According to the method, the stress concentration phenomenon of a joint elastomer can be effectively reduced, the lightness and thinness of the structure of the joint elastomer can be improved, and the impact energy can be stored and released, so that the dynamic stability and the load capacity of the joint are remarkably enhanced while the stress uniformity and the rigidity coefficient improvement are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of robot joint elastic body optimization, and in particular to a joint elastic body optimization design method based on a Gaussian filtering algorithm. Background Art

[0002] In the field of modern robotics, the performance of joint elastomers plays a vital role in the movement efficiency and stability of the entire robot. Thanks to the rapid development of modern information technologies such as computer technology and artificial intelligence technology, intelligent robots are gradually replacing humans and are widely used in complex and dangerous tasks such as disaster relief, terrain exploration, and material transportation. These tasks often require a high degree of dynamic stability in a complex and changing environment, but the traditional design method of joint elastomers fails to fully consider the handling of impact energy by the joints during robot movement, resulting in limited performance in actual use. In addition, when meeting high loads, the weight and volume of the robot structure will also increase, affecting the flexibility and response speed of the robot. Summary of the invention

[0003] In order to solve the above technical problems existing in the prior art, the present invention proposes a joint elastic body optimization design method based on Gaussian filtering algorithm. During the movement process, the joint elastic body can effectively store and release the impact energy, thereby significantly enhancing the dynamic stability of the robot system and improving its load capacity. The specific technical scheme is as follows:

[0004] A joint elastic body optimization design method based on Gaussian filtering algorithm, comprising:

[0005] Step S1, establishing a finite element simulation model of a joint elastic body based on an Archimedean spiral parameter equation;

[0006] Step S2, constructing an optimization evaluation function, which takes maximizing the deformation of the joint elastic body and minimizing its stress as the optimization goal;

[0007] Step S3, using the constructed optimization evaluation function as a constraint condition, selecting key design parameters of the elastomer for iterative optimization, the selected parameters including a bridge depth parameter H, which represents the depth of the H-bridge structure of the elastomer, and obtaining a threshold value of the optimization region of the H-bridge structure of the elastomer through iterative optimization;

[0008] Step S4, using a Gaussian filtering algorithm to smooth the H-bridge edge curve of the elastic body.

[0009] Furthermore, in step S1, the polar coordinate equation of the Archimedean spiral is:

[0010] r=a+b·θ

[0011] Among them, r is the distance from the origin to any point on the spiral line, a and b are real numbers; a is the distance between the starting point and the center of the polar coordinate when θ=0, b is used to control the distance between the spiral lines, representing the density of the spiral lines, θ is the angle parameter in the polar coordinates, and the larger θ is, the larger the range of the spiral line is;

[0012] The polar coordinate equation is converted into a Cartesian coordinate equation:

[0013] x=(α+β·θ)·cos(θ)

[0014] y=(α+β·θ)·sin(θ)

[0015] Among them, the parameters α and β are constants.

[0016] Furthermore, in step S2, the expression of the optimization evaluation function is as follows:

[0017] J(X)=w1·ε max (X)-w2·σ max (X)

[0018] Among them, ε max (X) is the maximum deformation of the elastic body when subjected to force, indicating the maximum deformation degree that the material can achieve under the condition of load X; σ max (X) is the maximum stress value per unit area of ​​the elastomer when subjected to force, reflecting the maximum stress level inside the material; w1 and w2 are coefficients used to balance the weights of the two optimization objectives.

[0019] Furthermore, in step S3, the deformation and stress of the elastic body increase with the increase of the bridge depth parameter H, affecting the local stiffness and load distribution of the elastic body; the key design parameters of the selected elastic body also include: the width parameter h, which represents the lateral dimension of the elastic body and directly affects its mechanical properties and space occupancy; the width variation parameter d n , which represents the change in the width of the elastic body along the direction of the helix, is used to adjust its elastic properties.

[0020] Furthermore, the optimization evaluation function is used as a constraint condition to adjust the width parameter h and the width variation parameter d. n , the bridge depth parameter H is iteratively optimized in stages. In each stage, finite element analysis is used to evaluate the current performance of the design parameters. According to the results of the optimization evaluation function and the real-time feedback mechanism to guide the parameter search direction, the key design parameters are updated and iterated to obtain the optimal solution.

[0021] Furthermore, the step S4 specifically includes:

[0022] Step S4.1: Obtain the threshold of the elastic body H-bridge structure optimization area in step S3, perform edge detection on the model point cloud data that is less than the set threshold, identify potential edge points, and extract edge curves;

[0023] Step S4.2: Establish a two-dimensional Gaussian function and initialize the Gaussian kernel;

[0024] Step S4.3: For each edge point in the edge curve, a two-dimensional Gaussian function is used for smoothing filtering to obtain a smooth edge curve.

[0025] Furthermore, in step S4.1, the edge point recognition method is:

[0026] Traverse each point cloud data, for each point P in the point cloud data i , check in P i Is there at least one point P in each of the four quadrants with δ as the center and the search radius? j (j≠i), the expression for judging whether it is an edge point is:

[0027]

[0028] Among them, isEdge=1 indicates that the point P i isEdge=0, which means that the point P i is a non-edge point.

[0029] Furthermore, in step S4.2, the expression of the two-dimensional Gaussian function is:

[0030]

[0031] Among them, σ represents the standard deviation of the Gaussian kernel, which is used to control the width of the Gaussian distribution and the smoothness of the curve.

[0032] Furthermore, the step S4.3 is specifically as follows: a Gaussian kernel constructed by a two-dimensional Gaussian function is applied to each edge point P of the point cloud data. i , and perform weighted averaging to obtain the point cloud coordinates P after Gaussian filtering i ', the calculation expression of weighted average is as follows:

[0033]

[0034] Among them, x i ' and y i 'It's point P i x and y after Gaussian filtering; m and n are the offsets relative to the center, K is the size of the Gaussian kernel; is the weight of the Gaussian kernel at offset (m,n).

[0035] The present invention realizes comprehensive and systematic optimization of the joint elastic body design by establishing an accurate simulation model and constructing a multi-objective optimization evaluation function; through an iterative optimization method and a Gaussian filtering algorithm, the width parameter h and the width variation parameter d of the elastic body are ensured. n The optimal configuration of these three design parameters, the bridge depth parameter H, and the smoothing of the edge curves significantly improve the quality of point cloud data processing and the reliability of the robot joints, ensuring that the design meets the performance standards in terms of dynamic stability and load capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic flow chart of a joint elasticity optimization method according to an embodiment of the present invention;

[0037] Figure 2 is a cross-sectional schematic diagram of an H-bridge structure according to an embodiment of the present invention;

[0038] Figure 3 is a comparison diagram of the edge curve before and after smoothing in an embodiment of the present invention;

[0039] Figure 4 It is a schematic diagram of the structure of the joint elastomer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and technical effect of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0041] like Figure 1 As shown, the present invention discloses a joint elastic body optimization design method based on Gaussian filtering algorithm, the steps comprising:

[0042] Step S1, designing a joint elastic body based on the Archimedean spiral: based on the Archimedean spiral parameter equation, establishing a finite element simulation model of the joint elastic body.

[0043] Specifically, the polar coordinate equation of the Archimedean spiral is:

[0044] r=a+b·θ

[0045] Where a and b are both real numbers; a is the distance between the starting point and the center of the polar coordinates when θ=0, It is used to control the distance between spiral lines, representing the density of the spiral lines. θ is the angle parameter in polar coordinates. The larger θ is, the larger the range of the spiral lines is.

[0046] The plane Cartesian coordinate equation of the Archimedean spiral is:

[0047] x=(α+β·θ)·cos(θ)

[0048] y=(α+β·θ)·sin(θ)

[0049] Among them, parameters α and β are constants. The changes of parameters α and β will affect the shape and spacing of the spiral. α changes the starting position of the spiral, while β controls the distance between the spirals.

[0050] In this embodiment, the inner helix of the elastic body simulation model has α1=9.2, β1=3; the outer helix has α2=16, β2=3.

[0051] The material of the joint elastic body is selected from maraging steel 300, and its performance parameters are shown in Table 1. The maximum load of the joint elastic body is T = 50N·m, when loaded:

[0052] stε max ≥3mm

[0053] σ max ≤1200Mpa.

[0054]

[0055] Table 1 Performance parameters of maraging steel 300.

[0056] Step S2, constructing an elastic body optimization function with the goal of maximizing deformation and minimizing stress: constructing a multi-objective optimization evaluation function, which has the optimization goals of maximizing the deformation of the joint elastic body and minimizing its stress, to ensure that the impact energy can be effectively stored and released during the movement.

[0057] Specifically, the expression of the optimization evaluation function is as follows:

[0058] J(X)=w1·ε max (X)-w2·σ max (X)

[0059] Among them, ε max (X) is the maximum deformation of the elastic body when subjected to force, indicating the maximum deformation degree that the material can achieve under the condition of load X; σ max (X) is the maximum stress value per unit area of ​​the elastic body when subjected to force, reflecting the maximum stress level inside the material; w1 and w2 are coefficients used to balance the weights of the two optimization objectives, and the values ​​of w1 and w2 are in Under the conditions, satisfy And are both greater than zero.

[0060] Step S3, the key design parameters of the elastomer are selected for iterative optimization: the design parameters of the elastomer to be optimized are determined, and the constructed multi-objective optimization evaluation function is used as a constraint condition, and the design parameters are systematically adjusted by iterative optimization until the predetermined performance standards are met to achieve the optimal configuration of the parameters.

[0061] Specifically, three main optimized design parameters of the elastic body are determined, which are the width parameter, width variation parameter and bridge depth parameter of the elastic body.

[0062] Width parameter h: determines the lateral dimension of the elastomer, directly affecting its mechanical properties and space occupancy;

[0063] Width variation parameter d n : Involves the change in the width of the elastic body along the helical direction, which is used to adjust its elastic properties;

[0064] Bridge depth parameter H: It is related to the depth of the H-bridge structure. The deformation and stress of the elastic body increase with the increase of H, which has a significant impact on the local stiffness and load distribution of the elastic body. The H-bridge structure is as follows: Figure 2 shown.

[0065] Using the evaluation function in step S2 as a constraint, the three target parameters are iteratively optimized in stages. Each step uses finite element analysis to evaluate the current performance of the design parameters, and the design parameters are updated according to the results of the evaluation function. The optimization method guides the parameter search direction through a real-time feedback mechanism to ensure that each iteration can approach the global optimal solution. The optimal parameter configuration is finally taken as: h = 2, H=1.2mm.

[0066] Step S4, smoothing the edge curve of the H-bridge of the elastic body: determine the threshold of the H-bridge edge, use the Gaussian filtering algorithm to smooth the edge curve of the H-bridge, and select the appropriate Gaussian kernel size and standard deviation to achieve the best smoothing effect. Under the premise of ensuring that the elastic body is evenly stressed, stress concentration is reduced, and the stiffness coefficient of the elastic body is improved, the structural design is further optimized to achieve a thinner structure and improve the dynamic stability and load capacity during joint movement. Specifically includes the following sub-steps:

[0067] Step S4.1, performing edge detection on the point cloud data that is smaller than a set threshold, identifying potential edge points, and extracting edge curves;

[0068] Determine the threshold σ of the elastic body H-bridge optimization region based on step S3 th =200Mpa, P i is the i-th point in the point cloud data, δ is the preset search radius, and N is the total number of point sets. i , check in P iIs there at least one point P in each of the four quadrants with ? as the center and δ as the radius? j (j≠i), the judgment expression is:

[0069]

[0070] Among them, isEdge=1 indicates that the point P i isEdge=0, which means that the point P i is a non-edge point.

[0071] Step S4.2, establishing a two-dimensional Gaussian function and initializing the Gaussian kernel;

[0072] The expression of the two-dimensional Gaussian function is:

[0073]

[0074] Among them, the standard deviation of the Gaussian kernel σ=2 controls the width of the Gaussian distribution and the smoothness of the curve.

[0075] Step S4.3: for each edge point P in the edge curve obtained above i , a two-dimensional Gaussian function is used for smoothing filtering to reduce noise and improve the continuity of edge curves;

[0076] The Gaussian kernel constructed by the two-dimensional Gaussian function is applied to each edge point P of the point cloud data i , weighted averaging is performed to achieve smoothing of the point cloud data. The point cloud coordinates P after Gaussian filtering are i 'Calculated by the following weighted average formula:

[0077]

[0078] Among them, x i ' and y i 'It's point P i x and y after Gaussian filtering; m and n are the offsets relative to the center, K = 5 is the size of the Gaussian kernel; is the weight of the Gaussian kernel at the offset (m,n). The weight G is determined by the Gaussian function, ensuring that the contribution of the neighboring points to the filtered coordinates is consistent with their distance P. i The closer the object is, the greater the weight.

[0079] Through the above steps, a smoothed edge curve is finally obtained, such as Figure 3 This process not only retains the key features of the edge points, but also improves the overall quality of the curve. Finally, the final joint elastic body is optimized, as shown in Figure 4As shown in the figure. Through this series of improvements, it is possible to optimize the stress distribution of the elastic body while ensuring the ideal mechanical response of the elastic body, and improve the storage and release of impact energy. Ultimately, it not only enhances the dynamic stability and load capacity of the robot joint, but also optimizes the cost-effectiveness.

[0080] The above is only a preferred implementation case of the present invention and does not limit the present invention in any form. Although the implementation process of the present invention is described in detail above, for those familiar with the art, they can still modify the technical solutions recorded in the above examples, or replace some of the technical features therein with equivalents. All modifications, equivalent replacements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A joint elastic body optimization design method based on Gaussian filtering algorithm, characterized in that: include: Step S1, establishing a finite element simulation model of a joint elastic body based on an Archimedean spiral parameter equation; Step S2, constructing an optimization evaluation function, which takes maximizing the deformation of the joint elastic body and minimizing its stress as the optimization goal; Step S3, using the constructed optimization evaluation function as a constraint condition, selecting key design parameters of the elastomer for iterative optimization, the selected parameters including a bridge depth parameter H, which represents the depth of the H-bridge structure of the elastomer, and obtaining a threshold value of the optimization region of the H-bridge structure of the elastomer through iterative optimization; Step S4, using a Gaussian filtering algorithm to smooth the H-bridge edge curve of the elastic body.

2. The joint elastic body optimization design method according to claim 1, characterized in that: In step S1, the polar coordinate equation of the Archimedean spiral is: r=a+b·θ Among them, r is the distance from the origin to any point on the spiral line, a and b are real numbers; a is the distance between the starting point and the center of the polar coordinate when θ=0, b is used to control the distance between the spiral lines, representing the density of the spiral lines, θ is the angle parameter in the polar coordinates, and the larger θ is, the larger the range of the spiral line is; The polar coordinate equation is converted into a Cartesian coordinate equation: x=(α+β·θ)·cos(θ) y=(α+β·θ)·sin(θ) Among them, the parameters α and β are constants.

3. The joint elastic body optimization design method according to claim 1, characterized in that: In step S2, the expression of the optimization evaluation function is as follows: J(X)=w1·ε max (X)-w2·σ max (X) Among them, ε max (X) is the maximum deformation of the elastic body when subjected to force, indicating the maximum deformation degree that the material can achieve under the condition of load X; σ max (X) is the maximum stress value per unit area of ​​the elastomer when subjected to force, reflecting the maximum stress level inside the material; w1 and w2 are coefficients used to balance the weights of the two optimization objectives.

4. The joint elastic body optimization design method according to claim 1, characterized in that: In step S3, the deformation and stress of the elastic body increase with the increase of the bridge depth parameter H, affecting the local stiffness and load distribution of the elastic body; the key design parameters of the selected elastic body also include: the width parameter h, which represents the lateral dimension of the elastic body and directly affects its mechanical properties and space occupancy; the width variation parameter d n , which represents the change in the width of the elastic body along the direction of the helix, is used to adjust its elastic properties.

5. The joint elastic body optimization design method according to claim 4, characterized in that: Taking the optimization evaluation function as the constraint condition, the width parameter h and the width change parameter d n , the bridge depth parameter H is iteratively optimized in stages. In each stage, finite element analysis is used to evaluate the current performance of the design parameters. According to the results of the optimization evaluation function and the real-time feedback mechanism to guide the parameter search direction, the key design parameters are updated and iterated to obtain the optimal solution.

6. The joint elastic body optimization design method according to claim 1, characterized in that: The step S4 specifically includes: Step S4.1: Obtain the threshold of the elastic body H-bridge structure optimization area in step S3, perform edge detection on the model point cloud data that is less than the set threshold, identify potential edge points, and extract edge curves; Step S4.2: Establish a two-dimensional Gaussian function and initialize the Gaussian kernel; Step S4.3: For each edge point in the edge curve, a two-dimensional Gaussian function is used for smoothing filtering to obtain a smooth edge curve.

7. The joint elastic body optimization design method according to claim 6, characterized in that: In step S4.1, the edge point identification method is: Traverse each point cloud data, for each point P in the point cloud data i , check in P i Is there at least one point P in each of the four quadrants with δ as the center and the search radius? j (j≠i), the expression for judging whether it is an edge point is: Among them, isEdge=1 indicates that the point P i isEdge=0, which means that the point P i is a non-edge point.

8. The joint elastic body optimization design method according to claim 7, characterized in that: In step S4.2, the expression of the two-dimensional Gaussian function is: Among them, σ represents the standard deviation of the Gaussian kernel, which is used to control the width of the Gaussian distribution and the smoothness of the curve.

9. The joint elastic body optimization design method according to claim 8, characterized in that: The step S4.3 is specifically: applying the Gaussian kernel constructed by the two-dimensional Gaussian function to each edge point P of the point cloud data. i , and perform weighted averaging to obtain the point cloud coordinates P after Gaussian filtering i ', the calculation expression of weighted average is as follows: Among them, x i ' and y i 'It's point P i x and y after Gaussian filtering; m and n are the offsets relative to the center, K is the size of the Gaussian kernel; is the weight of the Gaussian kernel at offset (m,n).

Images