A Robotic Repair Method for Automotive Paint Defects Based on Optimal Tool Deployment

Abstract

This application provides a robotic repair method for vehicle body paint defects based on optimal tool deployment. It utilizes known locations of paint defect points on a STL model of the vehicle surface. Based on the triangular mesh and corresponding normal vector of each defect point, defects that can be machined along the same tool axis direction are grouped into one category, and the corresponding tool axis direction is determined. Paint defect points along the same tool axis direction are projected onto a two-dimensional plane. The tool working dimensions are determined based on the machining tool's rotation radius and rosette machining trajectory. Based on the tool working dimensions and the projected positions of each paint defect point, the two-dimensional tool position requiring the fewest machining operations for defect repair along the same tool axis direction is determined using the optimal tool deployment method. The two-dimensional tool position is then inversely projected into three-dimensional space to determine the tool position's three-dimensional coordinates, and robot path planning and optimization are implemented based on this. This method can guide the robot to complete local defect repair, effectively improving the accuracy and efficiency of local robot machining.

Description

Technical Field

[0001] This application relates to the field of robotics, and more particularly to a robotic repair method for vehicle body paint defects based on optimal tool deployment. Background Technology

[0002] In the process of car body manufacturing, the repair of defects in the painting process is still a tedious manual process. Operators often need to work at close range, inspect the car body surface by visual inspection or touch, and then grind and polish away the defects in the car body paint. Due to the huge area of ​​the car body, there are problems such as low defect detection efficiency, high labor intensity of manual defect removal, and harsh working environment. At the same time, the instability of manual repair process will bring many challenges to the cost and quality control of car body manufacturers. In recent years, seeking to realize the automation, digitalization and intelligence of processing by robot technology has gradually become a new trend of efficient and high-quality manufacturing of complex car body parts. However, the random occurrence of car body paint defects puts forward strict requirements for the path planning of local repair robots. Through the above analysis, the main problems of the existing technology are: (1) Most of the local processing of parts can be pre-divided and positioned according to processing requirements or surface characteristics, which is difficult to apply to the local repair of randomly occurring paint defects; (2) The existing surface defect robot local repair is difficult to achieve high-efficiency and high-precision car body paint defect repair processing. Summary of the Invention

[0003] This application provides a robotic repair method for vehicle body paint defects based on optimal tool deployment. Based on the known locations of paint defect points on the STL model of the vehicle surface, defects that can be machined using the same tool axis direction are grouped together according to the triangular mesh and corresponding normal vector of each defect point, and the corresponding tool axis direction is determined. Paint defect points in the same tool axis direction are projected onto a two-dimensional plane along the tool axis direction. The tool working dimensions are determined based on the machining tool's rotation radius and rosette machining trajectory. Based on the tool working dimensions and the projected positions of each paint defect point, the two-dimensional tool position with the minimum number of machining operations required for defect repair in the same tool axis direction is determined using the optimal tool deployment method. The two-dimensional tool position is then inversely projected into three-dimensional space to determine the three-dimensional coordinates of the tool position, and robot path planning and optimization are implemented based on this.

[0004] The technical solution of this application is:

[0005] A robotic repair method for automotive paint defects based on optimal tool deployment includes the following steps:

[0006] A1, based on the triangular mesh and corresponding normal vector of each defect point, defects processed in the same tool axis direction are classified into one category;

[0007] A2, determine the tool axis direction required for machining defects with the same tool axis direction based on the triangular mesh and corresponding normal vector of each defect point;

[0008] A3, Project the defect points under the same tool axis direction onto a two-dimensional plane passing through the origin along the corresponding tool axis direction;

[0009] A4. Determine the tool working dimensions based on the machining tool's turning radius and the rose curve machining trajectory;

[0010] A5. Based on the tool working dimensions, the projected positions of each defect point, and the optimal tool deployment method, determine the two-dimensional tool position with the minimum number of machining operations required to repair defects in the same tool axis direction.

[0011] A6, determine the triangular mesh where the two-dimensional tool position is located and project it into three-dimensional space to obtain the tool position coordinates;

[0012] A7 uses an ant colony algorithm to determine the shortest path the robot takes throughout the machining process and determines the transition path based on the actual tool position and tool axis direction.

[0013] As one technical solution of this application, the specific steps for classifying defects according to the triangular mesh and corresponding normal vector of each defect point in step A1 are as follows:

[0014] A11, based on the first defect point in the defect point set D. Triangular grid normal quantity The normal vectors of the triangular meshes where other defect points are located and Defects with an included angle less than angle θ are grouped into one category and placed in set D1. Defect points are removed from set D, and the repair of defects in set D1 is carried out using the same tool axis direction, as shown in the following formula:

[0015] ;

[0016] In the formula: Let be the normal vector corresponding to the i-th defect point in the defect set D. Let θ be the normal vector corresponding to the first defect point in the defect set D, and let θ be the allowable angular deviation when sharing the same tool axis direction. Let D be the first defect point in set D, and D1 be the set of defect points that share the same tool axis direction as the first defect point in set D.

[0017] A12, based on the remaining defect points in the defect point set D, will be compared with the first defect point in the defect point set D. Defects of the same type are placed in set D2, and defective points in set D2 are removed from set D.

[0018] A13. Repeat steps A11 to A12 to classify the other defect points in set D until set D is empty.

[0019] As one technical solution of this application, in step A2, the method for determining the required tool axis direction for machining defects under the same tool axis direction is as follows:

[0020] The tool axis direction is determined by the area-weighted average of the normal vectors of the triangular meshes containing all defect points in the same defect set Di under the i-th tool axis direction. The formula is as follows:

[0021] ;

[0022] In the formula: t is the total number of defects of type i, j is the j-th defect of type i, and A j Di Let be the area of ​​the triangular mesh containing the j-th defect of type i, and let Di be the defect set corresponding to type i. Let be the normal vector corresponding to the j-th defect of the i-th type of defect.

[0023] As a technical solution of this application, in step A3, when the tool axis direction =(n x n y n z ) in n x =0、n y =0 and n z When ≠0, the projection formula is as follows:

[0024] ;

[0025] In the formula: (X p Y p (x) represents the projected coordinates. p y p Before projection, the coordinates (x) p y p , z p The first two items in );

[0026] When the tool axis direction =(n x n y n z For other cases, the projection formula is as follows:

[0027] ;

[0028] In the formula: (X p Y p (n) represents the projected coordinates.x n y n z ) is the tool axis direction vector, (x p y p , z p ) represents the coordinates before projection.

[0029] As one technical solution of this application, the method for determining the tool working dimensions in step A4 is as follows:

[0030] The tool working dimensions are determined based on the tool's turning radius and the rose-shaped trajectory of the tool when machining a certain position. The formula is as follows:

[0031] ;

[0032] In the formula: R w Let 'a' be the tool working dimension, 'k' be the amplitude of the rose trajectory, and 'k' be the number of petals. t For tool dimensions;

[0033] The rose line trajectory is as follows:

[0034] ;

[0035] In the formula: r is the distance from the point on the rose trajectory to the origin in polar coordinates, a is the amplitude of the rose trajectory, k is the number of petals, and θ is the angle in polar coordinates.

[0036] As one technical solution of this application, in step A5, the specific steps for determining the two-dimensional tool position with the minimum number of machining operations required to repair defects in the same tool axis direction are as follows:

[0037] A51, use equidistant square grids to divide the projection plane into rectangles;

[0038] A52, the mesh is scanned row by row after being divided, and the optimal mesh that covers the most defects and has the largest sum of removal power transmitted to the defects is selected as the next tool position projection point; the formula for the sum of removal power is as follows:

[0039] ,in, ;

[0040] In the formula: P Remij The tool position is provided with the removal efficiency for the j-th defect it covers, where l is the distance from the defect projection point to the tool position projection point, and R is the tool position projection point. w Let be the tool working dimension, Rem be the distance function of the defect point, N be the total number of tool positions, and M be the total number of defect points contained in the i tool positions;

[0041] A53, after placing one tool position, the covered defects are removed from the defect set Di to prevent duplicate coverage;

[0042] A54. Repeat steps A52 to A53 until the defect set is empty.

[0043] As one technical solution of this application, the specific steps for determining the triangular mesh where the two-dimensional tool position is located in step A6 are as follows:

[0044] A61, obtain the triangular mesh containing all defect points that can be covered by the two-dimensional tool position;

[0045] A62, project all triangular meshes that share vertices and edges with the triangular mesh where the defect point is located onto the two-dimensional plane along the tool axis direction;

[0046] A63, the triangular mesh in which the position of the two-dimensional tool is located is determined by the area comparison method;

[0047] A64, projecting the two-dimensional tool position onto three-dimensional space to obtain the tool position coordinates:

[0048] Among them, when the tool axis direction =(n x n y n z ) in n x =0、n y =0 and n z When ≠0, the inverse projection formula is as follows:

[0049] ;

[0050] In the formula: (X t Y t (x) is the projection point of the tool position. t y t , z t (x) is the inverse projection point of the tool position. q y q , z q ) represents a vertex on the triangular mesh where the two-dimensional tool position determined in step A6 is located, and n t =(n xt n yt n zt ) is the normal vector of the mesh;

[0051] When the tool axis direction =(n x n y n z For other cases, the inverse projection formula is as follows:

[0052] ;

[0053] In the formula, (X t Y t (x) is the projection point of the tool position. t y t , z t (x) is the inverse projection point of the tool position. q y q , z q ) represents a vertex on the triangular mesh where the two-dimensional tool position determined in step A6 is located, and n t =(n xt n yt n zt ) is the normal vector of the mesh, (n x n y n z ) is the tool axis direction vector.

[0054] As one technical solution of this application, the specific steps for determining the robot's shortest path and the transition path in step A7 are as follows:

[0055] A71, based on the tool position, the tool entry point and tool retraction point are set at δ points in the opposite direction of the tool axis. Considering only the tool entry point and tool retraction point, the shortest theoretical path of the robot is determined based on the ant colony algorithm. The path is then expressed as:

[0056] ;

[0057] In the formula: o is the machine origin, e1 is the first tool entry point or tool retraction point, e i e represents the i-th tool entry point or tool retraction point. i+1 For the (i+1)th tool entry point or tool retraction point, e n This refers to the last tool entry point or tool retraction point.

[0058] A72, the robot's actual path is as follows: the tool first needs to reach the entry point, then the working tool is activated and moved to the tool position, the machining is completed around the rose line, and then it returns to the retraction point to complete the retraction and moves to the next entry point, retraction point or machine origin;

[0059] A73 further optimizes the robot path, including the following two cases:

[0060] When there are two tool positions that are close to each other in the same tool axis direction in the actual path, the two tool positions are directly connected. That is, when the distance between the two tool positions is less than 1.5 times the tool working radius, the path is optimized so that the robot can directly feed from the previous tool position to the next tool position without passing through the tool entry point or tool retraction point.

[0061] When there are two consecutive tool entry points or retraction points in the theoretical path with different tool axis directions, a first plane is determined based on the first tool entry point or the first retraction point and the corresponding tool axis direction, and a second plane is determined based on the second tool entry point or the second retraction point and the corresponding tool axis direction. A first straight line is determined by the intersection of the first plane and the second plane, and a second straight line is determined by the first tool entry point and the second tool entry point or by the first retraction point and the second retraction point. The intersection of the skew perpendicular line between the first straight line and the second straight line and the second straight line is determined as a transition point.

[0062] A74, determine the shortest robot path and the transition path.

[0063] The beneficial effects of this application are:

[0064] This application provides a robotic repair method for vehicle body paint defects based on optimal tool deployment. It utilizes known locations of paint defect points on a STL model of the vehicle surface. Based on the triangular mesh and corresponding normal vector of each defect point, defects that can be machined along the same tool axis direction are grouped into one category, and the corresponding tool axis direction is determined. Paint defect points along the same tool axis direction are projected onto a two-dimensional plane. The tool working dimensions are determined based on the machining tool's rotation radius and rosette machining trajectory. Based on the tool working dimensions and the projected positions of each paint defect point, the two-dimensional tool position requiring the minimum number of machining operations for defect repair along the same tool axis direction is determined using the optimal tool deployment method. The two-dimensional tool position is then inversely projected into three-dimensional space to determine the tool position's three-dimensional coordinates, and robot path planning and optimization are performed based on this. The method disclosed in this invention guides the robot to complete local defect repair based on the location results of vehicle body surface defects, effectively improving the accuracy and efficiency of local robot machining. Attached Figure Description

[0065] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show...

[0066] Some embodiments of the application should not be considered as limiting the scope. For those skilled in the art, other related drawings can be obtained from these drawings without any inventive effort.

[0067] Figure 1 A flowchart illustrating the robotic repair method for vehicle body paint defects based on optimal tool deployment, provided in an embodiment of this application;

[0068] Figure 2 This is a schematic diagram of the tool working dimensions provided in the embodiments of this application;

[0069] Figure 3This is a schematic diagram of robot path optimization provided in an embodiment of this application.

[0070] Icons: 201 - Tool rotation radius; 202 - Tool position; 203 - Rose curve trajectory; 204 - Tool working radius; 301 - First tool position in the same tool axis direction; 302 - Second tool position in the same tool axis direction; 303 - Approach or retraction point corresponding to the first tool position; 304 - Approach or retraction point corresponding to the second tool position; 305 - First approach or retraction point corresponding to the previous tool position in a different tool axis direction; 306 - Second approach or retraction point corresponding to the next tool position in a different tool axis direction; 307 - First plane; 308 - Second plane; 309 - First straight line; 310 - Second straight line; 311 - Common perpendicular of skew planes; 312 - Intersection of the common perpendicular of skew planes and the second straight line. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can typically be arranged and designed in various different configurations.

[0072] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0073] Example:

[0074] Please refer to Figure 1 (Refer to) Figure 2 and Figure 3 This embodiment provides a robotic repair method for vehicle body paint defects based on optimal tool deployment, which mainly includes the following steps:

[0075] A1. Based on the triangular mesh and corresponding normal vector of each defect point, defects that can be machined using the same tool axis direction are grouped into one category. The specific steps are as follows:

[0076] A11, based on the first defect point in the defect point set D. Triangular grid normal quantity The normal vectors of the triangular meshes where other defect points are located and Defects with an included angle less than angle θ are grouped into one category and placed in set D1. These defect points are removed from set D. Defect repair in set D1 can be achieved using the same tool axis direction, and the formula is as follows:

[0077] ;

[0078] In the formula: Let be the normal vector corresponding to the i-th defect point in the defect set D. Let θ be the normal vector corresponding to the first defect point in the defect set D, and let θ be the allowable angular deviation when sharing the same tool axis direction. Let D be the first defect point in set D, and D1 be the set of defect points that share the same tool axis direction as the first defect point in set D.

[0079] A12, based on the remaining defect points in the defect point set D, will be compared with the first defect point in the defect point set D. Defects of the same type are placed in set D2, and defective points in set D2 are removed from set D.

[0080] A13, repeat the operations in steps A11 to A12, based on the ideas of steps A11 and A12, continue to classify the defect points in set D until set D is empty;

[0081] A2. Based on the triangular mesh and corresponding normal vector of each defect point, determine the required tool axis direction for machining defects in the same tool axis direction. The specific method is as follows:

[0082] The tool axis direction is determined by the area-weighted average of the normal vectors of the triangular meshes containing all defect points in the same defect set Di under the i-th tool axis direction. The formula is as follows:

[0083] ;

[0084] In the formula: t is the total number of defects of type i, j is the j-th defect of type i, and A j Di Let be the area of ​​the triangular mesh containing the j-th defect of type i, and let Di be the defect set corresponding to type i. Let be the normal vector of the j-th defect of the i-th type of defect;

[0085] A3, Project the defect points along the same tool axis direction onto a two-dimensional plane passing through the origin:

[0086] When the tool axis direction =(n x n y n z ) in n x =0、ny =0 and n z When ≠0, the projection formula is as follows:

[0087] ;

[0088] In the formula: (X p Y p (x) represents the projected coordinates. p y p Before projection, the coordinates (x) p y p , z p The first two items in );

[0089] When the tool axis direction =(n x n y n z For other cases, the projection formula is as follows:

[0090] ;

[0091] In the formula: (X p Y p (n) represents the projected coordinates. x n y n z ) is the tool axis direction vector, (x p y p , z p () represents the coordinates before projection;

[0092] A4, determine the tool working dimensions based on the machining tool's turning radius of 201 and the rose curve machining path, such as... Figure 2 As shown, the method for determining the tool working dimensions is as follows:

[0093] The tool working dimension 204 is determined based on the tool turning radius 201 and the rose-shaped trajectory 203 of the tool's movement when machining a certain position 202; the formula is as follows:

[0094] ;

[0095] In the formula: R w Here, 'a' represents the tool working dimension, 'a' represents the amplitude of the rose trajectory 203, 'k' represents the number of petals, and 'R' represents the depth of the rose. t For tool dimensions;

[0096] The rose line trajectory 203 is as follows:

[0097] ;

[0098] In the formula: r is the distance from the point on the rose trajectory in polar coordinates to the origin, a is the amplitude of the rose trajectory 203, k is the number of petals, and θ is the angle in polar coordinates;

[0099] A5. Based on the tool working dimensions and the projected positions of each defect point, the two-dimensional tool position with the minimum number of machining operations required to repair defects in the same tool axis direction is determined using the optimal tool deployment method. The specific steps are as follows:

[0100] A51, use equidistant square grids to divide the projection plane into rectangles;

[0101] A52, the mesh is scanned row by row after generation, and the optimal mesh that covers the most defects and has the largest sum of removal power transmitted to the defects is selected as the next tool position projection point; the formula for the sum of removal power is as follows:

[0102] ,in, ;

[0103] In the formula: P Remij The tool position is provided with the removal efficiency for the j-th defect it covers, where l is the distance from the defect projection point to the tool position projection point, and R is the tool position projection point. w Let be the tool working dimension, Rem be the distance function of the defect point, N be the total number of tool positions, and M be the total number of defect points contained in the i tool positions;

[0104] A53, after placing one tool position, the covered defects are removed from the defect set Di to prevent duplicate coverage;

[0105] A54, repeat steps A52 to A53 until the defect set is empty;

[0106] A6. Determine the triangular mesh containing the 2D tool position and project it into 3D space to obtain the tool position coordinates. The specific steps are as follows:

[0107] A61, obtain the triangular mesh containing all defect points that can be covered by the two-dimensional tool position;

[0108] A62, Project all triangular meshes that share vertices and edges with the triangular mesh where the defect point is located onto the two-dimensional plane along the tool axis direction;

[0109] A63, the triangular mesh in which the position of the two-dimensional tool is located is determined by the area comparison method;

[0110] A64, projecting the two-dimensional tool position onto three-dimensional space to obtain the tool position coordinates:

[0111] Among them, when the tool axis direction =(n x ny n z ) in n x =0、n y =0 and n z When ≠0, the inverse projection formula is as follows:

[0112] ;

[0113] In the formula: (X t Y t (x) is the projection point of the tool position. t y t , z t (x) is the inverse projection point of the tool position. q y q , z q ) represents a vertex on the triangular mesh where the two-dimensional tool position determined in step A6 is located, and n t =(n xt n yt n zt ) is the normal vector of the mesh;

[0114] When the tool axis direction =(n x n y n z For other cases, the inverse projection formula is as follows:

[0115] ;

[0116] In the formula, (X t Y t (x) is the projection point of the tool position. t y t , z t (x) is the inverse projection point of the tool position. q y q , z q ) represents a vertex on the triangular mesh where the two-dimensional tool position determined in step A6 is located, and n t =(n xt n yt n zt ) is the normal vector of the mesh, (n x n y n z ) represents the tool axis direction vector;

[0117] A7. The shortest path traversed by the robot throughout the entire machining process is determined using the ant colony algorithm, and the transition path is determined based on the actual tool position 202 and the tool axis direction; the specific steps are as follows:

[0118] A71, based on tool position 202, the tool entry point and retraction point are set at δ in the opposite direction of the tool axis. Considering only the entry point and retraction point, the shortest theoretical path of the robot is determined based on the ant colony algorithm; the path can be represented as:

[0119] ;

[0120] In the formula: o is the machine origin, e1 is the first tool entry point or tool retraction point, e i e represents the i-th tool entry point or tool retraction point. i+1 For the (i+1)th tool entry point or tool retraction point, e n This refers to the last tool entry point or tool retraction point.

[0121] A72, the robot's actual path is as follows: the tool first needs to reach the entry point, then start the working tool and move to the tool position 202, complete the machining around the rose line, return to the retraction point to complete the retraction, and move to the next entry point, retraction point or machine origin;

[0122] For A73, further optimization of the robot path requires consideration of the following two scenarios:

[0123] (1) When there are two closely spaced first tool positions 301 and second tool positions 302 in the same tool axis direction in the actual path, the first tool position 301 and the second tool position 302 are directly connected. That is, when the distance between the two tool positions is less than 1.5 times the tool working radius 204, the path is optimized so that the robot can directly feed from the previous tool position 301 to the next tool position 302 without passing through the tool entry point or tool exit point 303 corresponding to the first tool position and the tool entry point or tool exit point 304 corresponding to the second tool position.

[0124] (2) When there are two consecutive tool entry points or retraction points in the theoretical path with different tool axis directions, the first plane 307 is determined according to the three-dimensional coordinates of the two consecutive first tool entry points or first retraction points 305 and their corresponding tool axis directions, the second plane 308 is determined according to the second tool entry point or second retraction point 306 and its corresponding tool axis direction, the first straight line 309 is determined by the intersection of the first plane 307 and the second plane 308, the second straight line 310 is determined by the first tool entry point 305 and the second tool entry point 306 or by the first retraction point 305 and the second retraction point 306, and the intersection of the skew perpendicular line 311 between the first straight line 309 and the second straight line 310 and the second straight line 310 is determined as the transition point;

[0125] A74, the robot path needs to be determined based on the actual operation conditions in steps A71 to A73.

[0126] In summary, this application provides a robotic repair method for vehicle body paint defects based on optimal tool deployment. It discloses a method for localized robotic repair of vehicle body paint defects in the field of intelligent robotic machining. Furthermore, by guiding the robot to complete defect repair based on the vehicle surface defect location results, it achieves high efficiency and high precision while meeting processing requirements. Based on the known locations of paint defect points on the STL model of the vehicle surface, defects that can be processed using the same tool axis direction are grouped into one category and their corresponding tool axis directions are determined according to the triangular mesh and corresponding normal vector of each defect point. Paint defect points in the same tool axis direction are projected onto a two-dimensional plane along the tool axis direction. The tool working size is determined based on the machining tool's rotation radius 201 and the rose curve machining trajectory. Based on the tool working size and the projected positions of each paint defect point, the two-dimensional tool position with the minimum number of machining operations required for defect repair in the same tool axis direction is determined using the optimal tool deployment method. The two-dimensional tool position is then inversely projected into three-dimensional space to determine the three-dimensional coordinates of the tool position, and robot path planning and optimization are implemented based on this. The method disclosed in this invention guides a robot to complete local repair of defects based on the location results of defects on the vehicle body surface, which can effectively improve the accuracy and efficiency of robot local processing.

[0127] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A robot repair method for vehicle body paint defects based on optimal tool deployment, characterized by, The method comprises the following steps: A1, defects processed in the same tool axis direction are classified into one category according to the triangular mesh where each defect point is located and the corresponding normal vector; in step A1, the specific steps of classifying defects according to the triangular mesh where each defect point is located and the corresponding normal vector are as follows: A11, based on the first defect point in the defect point set D The normal vector of the triangular mesh where the defect point is located The normal vector of the triangular mesh where the defect point is located The normal vector of the triangular mesh where the defect point is located The normal vector of the triangular mesh where the defect point is located The defects with the angle between the normal vectors of the triangular meshes where the defects are located less than the angle θ are classified into one category and placed in the set D1, the defect points are removed from the set D, and the defects in the set D1 are repaired using the same tool axis direction, and the formula is as follows: ； In the formula: is the normal vector corresponding to the i-th defect point in the defect set D, is the normal vector corresponding to the first defect point in the defect set D, and θ is the allowable angular deviation in the same tool axis direction, is the first defect point in the set D, and D1 is the defect point set sharing the same tool axis direction with the first defect point in the set D; A12, based on the remaining defect points in the set D, will be placed in the set D2 with the first defect point in the set D Defect points of the same kind are placed in the set D2, while the defect points in the set D2 are removed from the set D; A13, repeat the operations in steps A11 to A12 to classify other defect points in the set D until the defect points in the set D are empty; A2, determine the tool axis direction required for processing defects in the same tool axis direction according to the triangular mesh where each defect point is located and the corresponding normal vector; A3, project the defect points in the same tool axis direction onto a two-dimensional plane along the corresponding tool axis direction; A4, determine the tool operation size according to the tool rotation radius and the rose curve machining trajectory; in step A4, the method for determining the tool operation size is as follows: Determine the tool operation size according to the tool rotation radius and the rose curve trajectory when the tool processes a certain position, and the formula is as follows: ； wherein: R w is the tool working dimension, a is the amplitude of the rose curve trajectory, k is the number of petals, R t is the tool dimension; The rose curve trajectory is as follows: ； In the formula, r is the distance from the rose curve trajectory point to the origin in polar coordinates, a is the amplitude of the rose curve trajectory, k is the number of petals, and θ is the angle in polar coordinates; A5, determine the two-dimensional tool position required for repairing defects in the same tool axis direction with the minimum number of machining times according to the tool operation size, the projection position of each defect point, and the optimal tool deployment method; A6, determine the triangular mesh where the two-dimensional tool position is located and project it into a three-dimensional space to obtain the tool position coordinates; A7, determine the shortest path of the robot in the entire machining process by using the ant colony algorithm, and determine the transition path according to the actual tool position and the tool axis direction.

2. The robotically repaired vehicle body paint defects based on optimal tool deployment method as claimed in claim 1, wherein, In step A2, the method for determining the tool axis direction required for processing defects in the same tool axis direction is as follows: The tool axis direction is determined according to the area-weighted average of the normal vectors of the triangular meshes of all defect points in the same defect set Di in the i-th tool axis direction , and the formula is as follows: ； where t is the total number of defects of the i-th type, j is the j-th defect of the i-th type, A j Di is the area of the triangle mesh where the j-th defect of the i-th type is located, Di is the defect set corresponding to the i-th type of defects, is the corresponding normal vector of the j-th defect of the i-th type.

3. The robotically repaired vehicle body paint defects based on optimal tool deployment method as claimed in claim 1, wherein, In step A3, when the tool axis direction = (n x , n y , n z ) with n x = 0, n y = 0 and n z ≠ 0, the projection formula is as follows: ； where: (X p , Y p ) are the projected coordinates, x p , y p are the first two terms in the pre-projected coordinates (x p , y p , z p ); When the direction of the tool axis = (n x , n y , n z ) is otherwise, the projection formula is as follows: ； where: (X p , Y p ) are the projected coordinates, (n x , n y , n z ) are the tool axis direction vectors, and (x p , y p , z p ) are the pre-projection coordinates.

4. The robotically repaired vehicle body paint defects based on optimal tool deployment method as claimed in claim 1, wherein, In step A5, the specific steps of determining the two-dimensional tool position required for repairing defects in the same tool axis direction with the minimum number of machining times are as follows: A51, use equidistant square grids to divide the projection plane into rectangles; A52, scan the divided grids row by row, and select the optimal grid that can cover the most defects and has the maximum total removal power delivered to the defects as the next tool position projection point position; the formula for the total removal power is as follows: wherein ; wherein: P Remij is the removal efficiency of the jth defect covered by the ith tool position, l is the distance from the defect point projection to the tool position projection, R w is the tool working size, Rem is the distance function of the defect point, N is the total number of all tool positions, and M is the total number of defect points covered by the ith tool position. A53, when one tool position is placed, remove the covered defects from the defect set Di to prevent repeated coverage; A54, repeat steps A52 to A53 until the defect set is empty.

5. The robotically repaired vehicle body paint defects based on optimal tool deployment method as claimed in claim 1, wherein, In step A6, the specific steps of determining the triangular mesh where the two-dimensional tool position is located are as follows: A61, obtain the triangular mesh where all defect points covered by the two-dimensional tool position are located; A62, project all triangular meshes that share vertices and edges with the triangular mesh where the defect point is located onto a two-dimensional plane along the tool axis direction; A63, determine the triangular mesh where the two-dimensional tool position is located by using the area comparison method; A64, project the two-dimensional tool position into a three-dimensional space to obtain the tool position coordinates: wherein, when the direction of the tool axis = (n x , n y , n z ) with n x = 0, n y = 0 and n z ≠ 0, the inverse projection formula is as follows: ； wherein: (x t , y t ) is the tool position projection point, (x t , y t , z t ) is the tool position inverse projection point, (x q , y q , z q ) is a vertex of the triangular mesh on which the two-dimensional tool position determined in step A6 is located, n t = (n xt , n yt , n zt ) is the normal vector of the mesh; When the direction of the tool axis = (n x , n y , n z ) is otherwise, the back projection formula is as follows: ； where (X t , Y t ) is the tool position projection point, (x t , y t , z t ) is the tool position inverse projection point, (x q , y q , z q ) is a vertex on the triangular mesh where the two-dimensional tool position determined in step A6 is located, n t = (n xt , n yt , n zt ) is the normal vector of the mesh, and (n x , n y , n z ) is the tool axis direction vector.

6. The robotically repaired vehicle body paint defects based on optimal tool deployment method as claimed in claim 1, wherein, In step A7, the specific steps of determining the shortest path of the robot and determining the transition path are as follows: A71, according to the tool position, set the tool feed point and the tool retreat point along the reverse direction of the tool axis at δ, in the case of only considering the tool feed point and the tool retreat point, determine the shortest theoretical path of the robot based on the ant colony algorithm, and the path is represented as: ； wherein: o is the mechanical origin, ei is the first tool engagement point or tool disengagement point, e i is the i-th tool engagement point or tool disengagement point, e i+1 is the i+1-th tool engagement point or tool disengagement point, e n is the last tool engagement point or tool disengagement point; A72, the actual path of the robot is as follows: the tool first needs to reach the feed point, then the working tool is started and moved to the tool position, the rose curve is completed after machining, and then the tool is returned to the retreat point to complete the retreat and moved to the next feed point, retreat point or mechanical origin; A73, further optimize the robot path, including the following two cases: When there are two tool positions with similar distances in the same tool axis direction in the actual path, directly connect the two tool positions, that is, when the distance between the two tool positions is less than 1.5 times the tool working radius, the path is optimized to the robot directly feeding from the previous tool position to the next tool position without passing through the feed point or the retreat point; When there are two continuous feed points or retreat points with different tool axis directions in the theoretical path, determine the first plane according to the first feed point or the first retreat point and the corresponding tool axis direction, determine the second plane according to the second feed point or the second retreat point and the corresponding tool axis direction, determine the first straight line by intersecting the first plane and the second plane, determine the second straight line by the first feed point and the second feed point or by the first retreat point and the second retreat point, and determine the intersection point of the second straight line and the common perpendicular line between the first straight line and the second straight line as the transition point; A74, determine the shortest robot path and determine the transition path.

Images