Robot high-priority subregion grinding and polishing path planning method and system for complex curved surface and storage medium

Abstract

The invention discloses a robot high-priority subregion grinding and polishing path planning method and system for a complex curved surface. The method comprises the following specific steps: S1, acquiring three-dimensional data of a complex curved surface, and selecting effective processing points; s2, processing effective points based on curved surface information to obtain an area center point, and determining a sub-area by considering the position and the normal distance; s3, defining a priority evaluation function, and arranging the sub-region processing sequence according to the size; s4, performing path planning on the sub-regions, and optimizing the path spacing by adopting a golden section method; and S5, the curved surface redundant material is updated in real time along with polishing, and sub-region and path planning is conducted again till the whole curved surface is polished. A new method is provided for robot grinding of the complex curved surface, real-time updating of the residual allowance of the curved surface material is achieved, subarea planning and path planning are continuously updated, key grinding and polishing are conducted on the key area of the complex curved surface, and the method has important significance on high-quality machining of the robot.

Description

Technical Field

[0001] The invention belongs to the field of robot control, and in particular relates to a robot high-priority sub-region grinding and polishing path planning method for complex surfaces. Background Art

[0002] Currently, part grinding and polishing are primarily performed manually or using CNC machines. However, manual grinding and polishing presents challenges such as high cost, low production efficiency, poor working environment, and inconsistent surface finish quality. CNC machine grinding and polishing also presents challenges such as frequent tool changes, poor machine flexibility, and difficulty machining complex curved surfaces. Compared to CNC machine grinding and polishing, industrial robot-based grinding and polishing systems have a wider range of applications, offering advantages such as a small footprint, low cost, strong adaptability, and portability. They can be mounted on mobile platforms to achieve a larger working range, meeting the requirements for machining freedom and range for grinding and polishing complex curved surfaces and the inner walls of workpiece cavities. Furthermore, robots offer greater flexibility than CNC machine tools, ensuring better consistency in grinding and polishing quality.

[0003] Appropriate grinding and polishing path planning can not only uniformly remove material and improve the local surface quality of the workpiece after grinding and polishing, but also enhance the accuracy of surface correction and grinding and polishing. However, complex surface structures are characterized by drastic changes in geometric features and uneven machining loads. Existing traditional path planning methods are unable to dynamically adjust to the geometric shape when dealing with the irregularities of complex surfaces. This lack of adaptability may result in under- or over-grinding of some surfaces. Furthermore, they cannot dynamically adjust the grinding and polishing strategy based on surface material redundancy and lack the ability to process local surface details. Summary of the Invention

[0004] The main purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a robot high-priority sub-region grinding and polishing path planning method for complex surfaces. By dividing the complex surface into regions and sorting their processing priorities, and planning the grinding and polishing path according to the priority sorting, the problem of insufficient processing of local features of complex surfaces in the prior art is solved, ensuring that the robot can perform global material removal on the complex surface along the designed path, and at the same time focus on processing areas with large material margins to ensure high-quality and high-efficiency processing requirements, providing an effective solution for the application of efficient robots in complex surface grinding and polishing tasks.

[0005] To achieve the above objectives, the technical solutions provided by the present invention are:

[0006] A robot high-priority sub-region grinding and polishing path planning method for complex surfaces, characterized by comprising:

[0007] Step S1: Obtain the three-dimensional coordinates, normal vector and initial material depth distribution of the complex surface, and select the effective points to be processed by setting the material depth threshold;

[0008] Step S2: For the valid points selected in step S1, a set of candidate sub-region center points is obtained based on the material depth and spatial distribution, and the neighborhood of the center point is determined by comprehensively considering the position distance and normal distance to obtain the sub-region information;

[0009] Step S3: A material removal priority evaluation function is constructed by defining a material reward function, a motion cost function, and a spatial balance term, and a processing order is defined for each sub-region according to the size of the evaluation function;

[0010] Step S4: Calculate the direction of maximum material coverage in the sub-area, and generate a set of parallel paths covering the sub-area based on this direction. Optimize the path spacing using the golden section method, and optimize the dwell time of the path points.

[0011] Step S5: According to the priority sub-area planning, the sub-areas are sorted in sequence for grinding and polishing. After the first grinding and polishing is completed, the material redundancy of the entire complex surface is iteratively updated. Steps S1-S4 are repeated until the material redundancy of the entire surface meets the requirements. Grinding and polishing is stopped to complete the global material removal.

[0012] Preferably, the step S1 adopts: in a complex surface, let the surface point set be The corresponding normal vector is The material depth distribution is Given a material depth threshold τ d , then the valid point index set for:

[0013]

[0014] The valid point set is And the corresponding normal vector set N valid and material depth set D valid .

[0015] Preferably, the step S2 adopts the following method: based on the valid point set, M points with the largest material depth are selected:

[0016] C material ={i∈P valid |d i ≥d (k)},M=min{50,|P valid |}

[0017] Where, d (k) is the Mth largest value after depth sorting;

[0018] Mesh the complex surface, and the size of each mesh is:

[0019]

[0020] Where x max , x min ,y max ,y min is the boundary coordinate of the valid point in the XY plane;

[0021] For any valid point p i =(x i ,y i , z i ), the grid index to which it belongs

[0022]

[0023] Select the point with the maximum material depth in the grid as the candidate center point

[0024]

[0025] If there is no valid point in the grid, that is, Then the grid does not generate candidate points;

[0026] Get the candidate center point set C:

[0027]

[0028] For the candidate center point c (coordinates p c , the normal vector is n c ), calculate other valid points j (coordinates are P j , the normal vector is n j ) to its combined distance δ j for:

[0029]

[0030] Where w d is the position distance weight, w n is the normal distance weight, ||p j -p c || is the Euclidean distance between point j and point c, is the angle between the two normal vectors;

[0031] According to the combined distance, the points within a certain range around the center point are selected by defining the sub-region radius R to form a sub-region S. c :

[0032] S c={j∈P valid |δ j ≤R}

[0033] Preferably, the step S3 adopts: first, define the material reward function R c Evaluate the processing value of sub-region c:

[0034]

[0035] Where, is the Gaussian weight, is the maximum material depth within the sub-area;

[0036] Secondly, define the motion cost function C c Evaluate the motion cost required to process sub-region c:

[0037]

[0038] Where, ||p c || is the distance from the center point of the sub-region to the coordinate origin, is the angle between the center point of the sub-region and the Z axis;

[0039] Next, define the spatial balance term B c Balance the processing order of the left and right sides of the entire surface:

[0040]

[0041] Where x c is the X coordinate of the center point of the sub-region, x min and x max The minimum and maximum coordinates of the entire workpiece in the X direction;

[0042] Finally, considering the material reward function, motion cost function and spatial balance term, the priority evaluation function P is defined c :

[0043] P c =(R c -0.5×C c )+B c

[0044] Where R c It is the main driving factor, hoping to prioritize the processing of areas with high material removal value, and the motion cost C c Influence priority with negative weight, spatial balance item B c Its value is between ±0.1, and the priority is fine-tuned. c The size sorts the sub-region processing order.

[0045] Preferably, the step S4 adopts the following method: in the local coordinate system of the sub-region, a center point c is defined, two tangent vectors X′ and Y′ form an orthogonal basis of the tangent plane, and the direction vector can be expressed as d θ =cosθ·X'+sinθ·Y′,θ∈[0,π); for each point p in the subregion i The vector relative to the center point c is projected onto the tangent plane, and the local coordinates are q i =p i -c, then the projection of the point in the direction is proj i (θ)=q i ·d θ ;

[0046] For a given direction θ, the projection values ​​along that direction are calculated and sorted. The coverage of the material as the tool moves along that direction is then simulated. Specifically, for each point i (sorted by the projection values), the coverage area centered at that point and with a radius equal to the tool radius r is considered. On the projection values, the distance between adjacent points is the distance along the path. Therefore, the coverage C(θ) is defined as:

[0047]

[0048] Where m j is the material depth at point j, N i (θ)={j:|proj j (θ)-proj i (θ)|≤r} represents the set of points whose distance from point i in the projection direction does not exceed r;

[0049] Finding the optimal direction Maximize coverage and arrange parallel paths in this direction;

[0050] The golden section method is used to optimize the path spacing, and the optimal spacing s * for:

[0051]

[0052] Where, is the Gaussian covering function, σ = r / 3, r is the tool radius;

[0053] The number of paths is The Y coordinate of the path position is

[0054] For each path, generate equidistant points along the path direction Returning to the global coordinate system, the local path P of the sub-region g for:

[0055] P g=c+x j X+y k Y+0·Z

[0056] Where X, Y, and Z are the local coordinate base coordinate systems of the sub-region, and x = x max ′, x=x min ′, y=y max ′, y=y min ′ is the sub-region boundary;

[0057] For each path, the dwell time is optimized using the final least squares method:

[0058]

[0059] Where t is the dwell time vector to be optimized, A = η·W ji ·G ji is the influence matrix, which represents the material removal amount per unit residence time of the i-th path point on the j-th material point, b is the target material removal vector, [t min , t max ] is the residence time constraint.

[0060] Preferably, the step S5 adopts the following method: sorting the sub-regions according to their priority, grinding and polishing the sub-regions in sequence, and iteratively updating the material redundancy M of the entire complex surface after the first grinding and polishing is completed. updated (x,y):

[0061]

[0062] Where M0(x,y) is the initial remaining material depth distribution, M i (x, y) is the material removal amount at the path point;

[0063] Repeat steps S1-S4 until the redundancy of the entire surface material reaches the requirement, that is, M updated (x,y)≤0.01, the iteration stops and the global material removal is completed.

[0064] A robot high-priority sub-region grinding and polishing path planning system for complex surfaces, characterized by comprising:

[0065] The regional planning module is configured to process valid points based on material depth and spatial distribution to obtain a set of candidate sub-region center points, determine its neighborhood by comprehensively considering the position distance and normal distance of the center point, and obtain sub-region information;

[0066] The region sorting module is configured to define a material removal priority evaluation function by defining a material reward function, a motion cost function, and a spatial balance term, and define a processing order for each sub-region according to the size of the evaluation function;

[0067] a path planning module configured to calculate the direction of maximum material coverage within a sub-area, generate a set of parallel paths covering the sub-area based on this direction, optimize the path spacing using the golden section method, and optimize the dwell time of path points;

[0068] The iterative update module is configured to iteratively update the material redundancy of the entire complex surface, repeat steps S1-S4 until the material redundancy of the entire surface meets the requirement, stop planning grinding and polishing, and complete global material removal.

[0069] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, it can implement any method step described in claims 1 to 6 and call any functional module of the system described in claim 7.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. The robot's high-priority sub-region grinding and polishing path planning method designed for complex surfaces comprehensively considers local geometric features and residual material thickness, and can focus on grinding and polishing key areas of complex surfaces;

[0072] 2. The present invention can update the remaining allowance of complex surface materials in real time, and continuously update sub-area planning and path planning;

[0073] 3. The present invention can significantly reduce the peak error remaining after global processing and improve the processing quality of the robot for grinding and polishing complex surfaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to further explain the technical solution of the present invention, the present invention is briefly introduced below with reference to the drawings of specific embodiments.

[0075] Figure 1 A flow chart of a high-priority sub-region polishing path planning algorithm for complex surfaces provided by an embodiment of the present invention;

[0076] Figure 2 The complex surface three provided in the embodiment of the present invention is a model and initial material allowance distribution;

[0077] Figure 3 A simulation diagram of the cumulative polishing path of the first iteration is shown for an embodiment of the present invention;

[0078] Figure 4 A simulation diagram of the cumulative polishing path of the tenth iteration is shown in an embodiment of the present invention;

[0079] Figure 5 A simulation diagram of the cumulative polishing path of the twentieth iteration is shown in an embodiment of the present invention;

[0080] Figure 6 A schematic diagram comparing the polishing effects of a high-priority sub-region path planning method according to an embodiment of the present invention;

[0081] Figure 7 A schematic diagram comparing the polishing effects of the uniform grid scanning path planning method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0082] The following describes the implementation of the present invention through specific embodiments and in conjunction with the accompanying drawings.

[0083] In view of the obvious deficiencies of traditional path planning methods in dealing with surface irregularities and local detail processing, the present invention provides a robot high-priority sub-region grinding and polishing path planning method for complex surfaces. The complex surfaces are divided and path planning and polishing are performed according to regional priorities, which can significantly improve the grinding and polishing quality of complex surfaces.

[0084] Figure 1 The flowchart of the algorithm for planning the polishing path based on high-priority sub-regions for complex surfaces provided by the present invention includes the following steps:

[0085] Step S1: In a complex surface, let the surface point set be The corresponding normal vector is The material depth distribution is Given a material depth threshold τ d , then the valid point index set for:

[0086]

[0087] The valid point set is And the corresponding normal vector set N valid and material depth set D valid .

[0088] Step S2: Based on the valid point set, select M points with the largest material depth:

[0089] C material ={i∈P valid |d i ≥d (k)}, M = min{50, |P valid |}

[0090] Where, d (k) is the Mth largest value after depth sorting;

[0091] Mesh the complex surface, and the size of each mesh is:

[0092] G x =5, G y =5

[0093] Where x max , x min ,y max ,y min is the boundary coordinate of the valid point in the XY plane;

[0094] For any valid point p i =(x i ,y i , z i ), the grid index to which it belongs

[0095]

[0096] Select the point with the maximum material depth in the grid as the candidate center point

[0097]

[0098] If there is no valid point in the grid, that is, Then the grid does not generate candidate points;

[0099] Get the candidate center point set C:

[0100]

[0101] For the candidate center point c (coordinates p c , the normal vector is n c ), calculate other valid points j (coordinates are p j , the normal vector is n j ) to its combined distance δ j for:

[0102]

[0103] Where w d is the position distance weight, w n is the normal distance weight, ||p j -p c || is the Euclidean distance between point j and point c, is the angle between the two normal vectors;

[0104] According to the combined distance, the points within a certain range around the center point are selected by defining the sub-region radius R to form a sub-region S. c :

[0105] S c ={j∈P valid|δ j ≤R}

[0106] Step S3: First, define the material reward function R c Evaluate the processing value of sub-region c:

[0107]

[0108] Where, is the Gaussian weight, is the maximum material depth within the sub-area;

[0109] Secondly, define the motion cost function C c Evaluate the motion cost required to process sub-region c:

[0110]

[0111] Where, ||p c || is the distance from the center point of the sub-region to the coordinate origin, is the angle between the center point of the sub-region and the Z axis;

[0112] Next, define the spatial balance term B c Balance the processing order of the left and right sides of the entire surface:

[0113]

[0114] Where x c is the X coordinate of the center point of the sub-region, x min and x max The minimum and maximum coordinates of the entire workpiece in the X direction;

[0115] Finally, considering the material reward function, motion cost function and spatial balance term, the priority evaluation function P is defined c :

[0116] P c =(R c -0.5×C c )+B c

[0117] Where R c It is the main driving factor, hoping to prioritize the processing of areas with high material removal value, and the motion cost C c Influence priority with negative weight, spatial balance item B c Its value is between ±0.1, and the priority is fine-tuned. c The size sorts the sub-region processing order.

[0118] Step S4: In the local coordinate system of the sub-region, define the center point c, the two tangent vectors X′ and Y′ form the orthogonal basis of the tangent plane, and the direction vector can be expressed as d θ =cosθ·X′+sinθ·Y′, θ∈[0,π); for each point p in the subregion i The vector relative to the center point c is projected onto the tangent plane, and the local coordinates are q i =p i -c, then the projection of the point in the direction is proj i (θ)=q i ·d θ ;

[0119] For a given direction θ, the projection values ​​along that direction are calculated and sorted. The coverage of the material as the tool moves along that direction is then simulated. Specifically, for each point i (sorted by the projection values), the coverage area centered at that point and with a radius equal to the tool radius r is considered. On the projection values, the distance between adjacent points is the distance along the path. Therefore, the coverage C(θ) is defined as:

[0120]

[0121] Where m j is the material depth at point j, N i (θ)={j:|proj j (θ)-proj i (θ)|≤r} represents the set of points whose distance from point i in the projection direction does not exceed r;

[0122] Finding the optimal direction Maximize coverage and arrange parallel paths in this direction;

[0123] The golden section method is used to optimize the path spacing, and the optimal spacing s * for:

[0124]

[0125] Where, is the Gaussian covering function, σ = r / 3, r is the tool radius;

[0126] The number of paths is The Y coordinate of the path position is

[0127] For each path, generate equidistant points along the path direction Returning to the global coordinate system, the local path P of the sub-region g for:

[0128] P g =c+xj X+y k Y+0·Z

[0129] Where X, Y, and Z are the local coordinate base coordinate systems of the sub-region, and x = x max ′, x=x min ′, y=y max ′, y=y min ′ is the sub-region boundary;

[0130] For each path, the dwell time is optimized using the final least squares method:

[0131]

[0132] Where t is the dwell time vector to be optimized, A = η·W ji ·G ji is the influence matrix, which represents the material removal amount per unit residence time of the i-th path point on the j-th material point, b is the target material removal vector, [t min , t max ] is the residence time constraint.

[0133] Step S5: sort the sub-regions according to their priority, and polish them in turn. After the first polishing is completed, iteratively update the material redundancy M of the entire complex surface. updated (x, y):

[0134]

[0135] Where M0(x,y) is the initial remaining material depth distribution, M i (x, y) is the material removal amount at the path point;

[0136] Repeat steps S1-S4 until the redundancy of the entire surface material reaches the requirement, that is, M npdated When (x, y)≤0.01, the iteration stops and the global material removal is completed.

[0137] In order to verify the feasibility of this method, the above method was simulated in the MATLAB programming environment.

[0138] Figure 2 The third model of the complex surface and the distribution of the initial material allowance. The contour lines on the right represent the initial thickness distribution of the material that needs to be removed. The black area indicates a thicker allowance, with the maximum being about 0.5 mm, and the white area indicates a thinner allowance.

[0139] Figure 3 This is the cumulative grinding and polishing path effect of the first iteration. During the first iteration of grinding, the algorithm prioritizes identifying the area with the largest material allowance and performs local path planning and grinding on it.

[0140] Figure 4 This is the cumulative grinding and polishing path effect of the tenth iteration. As the number of iterations increases, the initial priority area is polished, the algorithm recalculates the surface material allowance and continues to perform local path planning and polishing on other sub-areas.

[0141] Figure 5 This is the cumulative polishing path effect of the 20th iteration. After the 20th iteration, the initial high-residue areas of the surface have been repeatedly covered, and the accumulated paths have formed denser trajectory clusters in these areas. In the areas with less residuals, the paths are much sparser, and only necessary paths are passed through.

[0142] Figure 6 、 7 Comparing the polishing results of the inventive method with those of a uniform grid scanning path planning method, the inventive method prioritizes polishing key areas of complex curved surfaces, effectively removing nearly all of the white areas, leaving no large areas of material remaining. The maximum thickness of the remaining material is less than 0.02mm, and the removal error in most areas is within 0.01mm, achieving a surface that approaches the ideal shape. Reducing the peak error remaining after global processing significantly improves the polishing quality of complex curved surfaces.

[0143] The present invention has been described above in conjunction with preferred embodiments. The present invention may also have various other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A robot high-priority sub-region polishing path planning method for complex surfaces, characterized by: include: Step S1: Obtain the three-dimensional coordinates, normal vector and initial material depth distribution of the complex surface, and select the effective points to be processed by setting the material depth threshold; Step S2: For the valid points selected in step S1, a set of candidate sub-region center points is obtained based on the material depth and spatial distribution, and the neighborhood of the center point is determined by comprehensively considering the position distance and normal distance to obtain the sub-region information; Step S3: A material removal priority evaluation function is constructed by defining a material reward function, a motion cost function, and a spatial balance term, and a processing order is defined for each sub-region according to the size of the evaluation function; Step S4: Calculate the direction of maximum material coverage in the sub-area, and generate a set of parallel paths covering the sub-area based on this direction. Optimize the path spacing using the golden section method, and optimize the dwell time of the path points. Step S5: According to the priority sub-area planning, the sub-areas are sorted in sequence for grinding and polishing. After the first grinding and polishing is completed, the material redundancy of the entire complex surface is iteratively updated. Steps S1-S4 are repeated until the material redundancy of the entire surface meets the requirements. Grinding and polishing is stopped to complete the global material removal.

2. The robot high-priority sub-region search method for complex surfaces according to claim 1, characterized in that: The step S1 adopts: obtaining the three-dimensional coordinates of the complex surface Normal vector and initial material depth distribution And by setting the material depth threshold τ d Establish a valid point index set The effective point set that needs to be processed is And the corresponding normal vector set N valid and material depth set D valid .

3. The robot high-priority sub-region search method for complex surfaces according to claim 1, characterized in that: The step S2 adopts: for the valid point set, calculating the candidate point C obtained based on the material depth material and the candidate points obtained based on grid division Get the candidate center point set For the candidate center point c (coordinates p c , the normal vector is n c ), calculate other valid points j (coordinates are p j , the normal vector is n j ) to its combined distance δ j for: Where w d is the position distance weight, w n is the normal distance weight, ||p j -p c || is the Euclidean distance between point j and point c, is the angle between the two normal vectors; According to the combined distance, the points within a certain range around the center point are selected by defining the sub-region radius R to form a sub-region S. c : S c ={j∈P valid |d j ≤R}。 4. The robot high-priority sub-region search method for complex surfaces according to claim 1, characterized in that: The step S3 adopts: first, define the material reward function R c Evaluate the processing value of sub-region c: Where, is the Gaussian weight, is the maximum material depth within the sub-area; Secondly, define the motion cost function C c Evaluate the motion cost required to process sub-region c: Where, ||p c || is the distance from the center point of the sub-region to the coordinate origin, is the angle between the center point of the sub-region and the Z axis; Next, define the spatial balance term B c Balance the processing order of the left and right sides of the entire surface: Where x c is the X coordinate of the center point of the sub-region, x min and x max The minimum and maximum coordinates of the entire workpiece in the X direction; Finally, considering the material reward function, motion cost function and spatial balance term, the material removal priority evaluation function P is defined c , sort the sub-region processing order according to the priority evaluation function size: P c =(R c -0.5×C c )+B c 。 5. The robot high-priority sub-region search method for complex surfaces according to claim 1, characterized in that: The step S4 adopts: for the sub-region, the coverage amount C(θ) is defined as: Where m j is the material depth at point j, N i (θ)={j:|proj j (θ)-proj i (θ)|≤r} represents the set of points whose distance from point i in the projection direction does not exceed r; and find the best direction Maximize coverage and arrange parallel paths in this direction; The golden section method is used to optimize the path spacing, and the optimal spacing s * for: Where, is the Gaussian covering function, σ=r / 3, r is the tool radius; Based on this, the number of paths can be obtained as The Y coordinate of the path position is For each path, generate equidistant points along the path direction Returning to the global coordinate system, the local path P of the sub-region g for: P g =c+x j X+y k Y+0·Z Where X, Y, and Z are the local coordinate base coordinate systems of the sub-region, and x = x max ′, x=x min ′, y=y max ′, y=y min ′ is the sub-region boundary; For each path, the dwell time is optimized using the final least squares method: Where t is the dwell time vector to be optimized, A = η·W ji ·G ji is the influence matrix, which represents the material removal amount per unit residence time of the i-th path point on the j-th material point, b is the target material removal vector, [t min , t max ] is the residence time constraint.

6. The robot high-priority sub-region search method for complex surfaces according to claim 1, characterized in that: The step S5 adopts the following steps: sorting the sub-regions according to their priority, grinding and polishing them in sequence, and iteratively updating the material redundancy M of the entire complex surface after the first grinding and polishing is completed. updated (x,y): Where M0(x, y) is the initial remaining material depth distribution, M i (x,y) is the material removal amount at the path point; According to the remaining material allowance, repeat the steps of sub-area planning, priority sub-area sorting, and local path planning until the entire surface material redundancy reaches the requirement, that is, M updated (x,y)≤0.01, the iteration stops and the global material removal is completed.

7. A robot high-priority sub-region polishing path planning system for complex surfaces, characterized by: include: The regional planning module is configured to process valid points based on material depth and spatial distribution to obtain a set of candidate sub-region center points, determine its neighborhood by comprehensively considering the position distance and normal distance of the center point, and obtain sub-region information; The region sorting module is configured to define a material removal priority evaluation function by defining a material reward function, a motion cost function, and a spatial balance term, and define a processing order for each sub-region according to the size of the evaluation function; a path planning module configured to calculate the direction of maximum material coverage within a sub-area, generate a set of parallel paths covering the sub-area based on this direction, optimize the path spacing using the golden section method, and optimize the dwell time of path points; The iterative update module is configured to iteratively update the material redundancy of the entire complex surface, repeat steps S1-S4 until the material redundancy of the entire surface meets the requirement, stop planning grinding and polishing, and complete global material removal.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it can implement any method step described in claims 1 to 6 and call any functional module of the system described in claim 7.

Images