Point cloud trajectory planning method for robot polishing system

Abstract

The invention discloses a robot polishing system-oriented point cloud trajectory planning method, which comprises the following steps of: establishing a robot polishing processing hardware system, and acquiring point cloud data of a curved surface composite component under a robot base mark; accurate and simplified point cloud data are obtained according to point cloud preprocessing, a point cloud processing algorithm is used for obtaining a machining pose micro-line segment, a pose optimization function is built according to robot pose constraint conditions, and an optimal machining process pose is obtained through a genetic algorithm; a position and posture transition model and speed constraint are established for prospective position and posture initialization, position and posture synchronous judgment conditions are set for determining information of a section to be lengthened, a T-shaped speed and time synchronization method is provided for position and posture synchronous adjustment and real-time machining track planning, and efficient grinding machining of the robot is achieved. According to the method, the processing pose is acquired and optimized according to the live point cloud, and pose synchronization look-ahead planning is realized; the problem that errors exist in a digital-analog generation track is solved, posture sudden change is avoided, the machining requirement is met, and the machining efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot polishing control, and in particular relates to a point cloud trajectory planning method for a robot polishing system. Background Art

[0002] Curved composite components are widely used in aerospace, automotive industry and other high-tech fields, and can provide high specific strength, corrosion resistance and designability. After the composite components are laid and formed, there are often surface wrinkles caused by vacuum guide pipes, sheet overlap and other operations. These wrinkles will affect the appearance, structural integrity and functional performance of the final product. Grinding and removing them is an important processing step in the post-processing of curved composite components.

[0003] In order to meet the grinding needs of complex curved surface structures of composite components, avoid the harm of grinding composite powder to workers, and improve the intelligence and automation of processing, industrial robots are also widely used in the field of industrial processing with their advantages of automation, flexibility, and high stability, and have a good development trend in the field of grinding and polishing.

[0004] Traditional robots generally obtain processing trajectories through offline programming, that is, obtaining processing micro-segments based on theoretical digital models. However, curved composite components cannot be accurately positioned, resulting in a mismatch between the actual situation and the digital model, so the offline programming method cannot meet actual processing requirements. However, point cloud data accurately records the actual three-dimensional information, which can solve the problem of deviation between the above-mentioned curved composite components and the theoretical digital model. In addition, in the robot posture synchronization planning, considering the frequent start and stop of the equipment due to the zero turning speed of the processing micro-segment and the problem of excessive posture changes due to the failure to consider the posture constraints of posture synchronization, it is necessary to perform smoother and more accurate trajectory planning for curved composite components. Summary of the invention

[0005] In response to the problems existing in the above-mentioned prior art, the present invention proposes a point cloud trajectory planning method for a robot polishing system, which obtains and optimizes the processing posture according to the real-time point cloud, determines the speed and time synchronization method, so as to realize the real-time forward-looking process of posture synchronization of robot analysis, solve the problem that the digital model cannot accurately locate, avoid sudden changes in posture, meet the surface polishing processing requirements of the robot system, and improve the robot polishing processing speed.

[0006] In order to achieve the above technical objectives, the present invention provides the following technical solutions:

[0007] A point cloud trajectory planning method for a robot polishing system, specifically comprising:

[0008] S1. Establish the hardware configuration of the industrial robot grinding system, including an industrial robot, a constant force flange, an eccentric grinder, a binocular vision measuring device, a curved composite component and a storage table; the constant force flange is fixedly connected to the eccentric grinder and installed at the end of the robot, responsible for grinding and controlling the processing force to be stable at a constant value; the curved composite component is placed on the storage table; the binocular vision measuring device is placed on the ground at a certain distance from the processing area to ensure that the curved composite component, that is, the component to be processed, is in a better position in the field of view;

[0009] S2. Use a binocular vision measuring device to establish a coordinate system, and use a spherical scanner in the binocular vision measuring device to obtain the three-dimensional point cloud information corresponding to the component to be processed;

[0010] S3, performing filtering sampling preprocessing on the three-dimensional point cloud information obtained in step S2, and obtaining the robot processing position and posture through a point cloud algorithm processing method;

[0011] S4, according to the processing posture obtained in step S3, using the robot processing posture constraint conditions, optimizing the robot processing posture, and obtaining the final robot posture optimization result;

[0012] S5. According to the posture result obtained in step S4, select N posture forward trajectory segments, N ≥ 3, and perform initialization parameter setting according to the posture transition model and speed constraints;

[0013] S6, obtaining the planning time of the straight line segment and arc segment of the robot position and posture through the T-type speed interpolation method, and performing position and posture synchronization judgment to determine the information of the lengthened segment to be synchronized;

[0014] S7, based on the posture synchronization judgment result of step S6, adjusting the posture synchronization of the arc segment and the straight line segment respectively by a speed synchronization method;

[0015] S8, and determine whether the last posture is synchronized forward, and make different data output and storage decisions; the output data plans the processing trajectory in real time according to the adjustment speed.

[0016] Furthermore, step S2 specifically includes:

[0017] The robot base coordinate system, workpiece coordinate system, wrist coordinate system and tool coordinate system are established by using binocular vision measuring equipment. The ball scan in the binocular vision measuring equipment is used to obtain the point cloud of the component to be processed relative to the workpiece coordinate system. Then, based on the conversion relationship between the workpiece coordinate system and the robot base coordinate system, the three-dimensional point cloud information of the component to be processed under the robot base coordinate system is obtained.

[0018] Furthermore, step S3 specifically includes:

[0019] S31. Use statistical filtering to remove noise data; perform statistics on the neighborhood of each point to obtain a result of approximate Gaussian distribution, and calculate the average distance d and standard deviation of each point to all adjacent points. σ , determine the standard deviation coefficient std, remove the points whose average distance is outside the standard range D = d + std × σ, that is, remove the points that are relatively "scattered" from the surrounding neighboring points;

[0020] S32, using the voxel downsampling method to obtain simplified point cloud data, determining rows, columns, and layers according to the maximum and minimum values ​​of the point cloud and the grid size, placing each point cloud data into the corresponding grid according to the coordinates, and calculating the centroid of the non-empty grid to replace all points in the voxel;

[0021] S33, using two planes separated by a distance d to intercept a point cloud layer, projecting the point cloud layer onto a middle plane between the two planes by projection, and obtaining initial path points; using a NURBS curve to fit the path points, and obtaining discrete processing positions with equidistant steps;

[0022] S34, using a point cloud normal vector algorithm, find a plane of point cloud fitting within a sphere of radius R of the processing position obtained in step S33, select any key point on one side of the curved surface, and obtain a normal vector of the fitting plane toward the key point as the normal vector of the processing position;

[0023] S35. Use the normal vector obtained in step S34 as the z-axis, and set the x and y axes arbitrarily while satisfying the xyz right-hand coordinate system principle to calculate the Euler angle of this xyz coordinate system relative to the robot base coordinate system; use the Euler angle and the coordinates of the processing position obtained in S33 as the processing posture; the robot base coordinate system is established by a binocular vision measurement device.

[0024] Furthermore, step S4 specifically includes:

[0025] S41. Calculate the joint angles corresponding to different robot terminal positions based on the robot inverse kinematics, obtain the joint range based on the robot technical parameters, and take the ratio of the joint position to the median of the joint range to the joint motion range as the robot joint index R. θ , as follows:

[0026]

[0027] in, n represents the number of robot joints, θ imin Indicates the minimum limit of the joint corresponding to joint i, 1≤i≤6, θ imax represents the maximum limit of the joint corresponding to joint i, θ imid Indicates that joint i corresponds to the middle limit of the joint, and θ i represents the angle of each joint, and θ imin≤θ i ≤θ imax ;

[0028] S42. Use maneuverability to comprehensively measure the robot's ability to move in all directions in a certain posture, and take the reciprocal of the maneuverability as the robot's dexterity index R. ω , the formula is:

[0029]

[0030] in, is the operability, J(θ) is the Jacobian matrix;

[0031] S43. On the basis of step S41 and step S42, dimensionless variables of robot joint index and robot dexterity index are obtained, and the robot end posture change angle corresponding to the processing path point is used as a variable, and a nonlinear objective function for obtaining the optimal value of the robot processing posture is established. A genetic algorithm is used to optimize and solve the target to obtain the optimal posture change angle; the obtained posture is used as the final robot processing posture.

[0032] Furthermore, step S5 specifically includes:

[0033] S51, perform position transition initialization; for the first forward initialization, obtain the arc radius under the maximum error, obtain half of the trajectory segment as the arc radius of the arc transition, take the smaller of the two radii as the initial value of the arc radius, and set the initial speed of the first forward segment and the terminal speed of the last segment to 0; the arc segment speed constraint is set according to the comprehensive restrictions of acceleration, normal acceleration, and maximum speed; the speed and acceleration constraints of the straight segment are respectively set according to the maximum speed and maximum acceleration restrictions; the connection point constraint conditions of the straight segment and the arc segment are set according to the arc segment speed and acceleration constraints;

[0034] S52, for initialization other than the first look-ahead, the look-ahead starting speed is set according to the final speed outputted by the last look-ahead, and the preset manner of the remaining variables is the same as the first look-ahead setting in step S51;

[0035] S53, on the basis of steps S51 and S52, the boundary speed is judged as acceleration / deceleration type according to the boundary speeds at both ends of the straight line, and a boundary speed determination method is formulated; during the acceleration process, the end speed constraint is reduced to the maximum speed to which the starting speed of the straight line segment can be accelerated; during the deceleration process, the starting speed constraint is reduced to the maximum speed to which the end speed of the straight line segment can be reversely accelerated, and the parameter setting of the position transition initialization is updated with this speed constraint;

[0036] S54. Similarly, perform posture transition initialization according to the process described in steps S51-S53.

[0037] More specifically, step S51 is as follows:

[0038] Note: n1 , P n2 , P n3 , P n4 , P n5 , P n6 , P n7 is the endpoint of the position micro segment, P0 is the endpoint of the position segment P n2 Corresponding to the center point of the transition arc, P1 and P2 are the tangent points of the position transition arc and the position straight line segment, P3 and P4 are the position points of two adjacent interpolation, and P5 is P0P n2 The intersection point with the arc, P6 is P3P4 and P0P n2 The intersection of P1P2 and P0P n2 The intersection of L1, L2, L3, L4, L5, L6 is the position segment P n1 P n2 , P n2 P n3 , P n3 P n4 , P n4 P n5 , P n5 P n6 , P n6 P n7 The corresponding length; note P n1 , P n2 , P n3 , P n4 , P n5 , P n6 , P n7 is the endpoint of the position micro segment, P0 is the endpoint of the position segment P n2 Corresponding to the center point of the transition arc segment, P1 and P2 are the tangent points of the position transition arc segment and the position straight line segment, P3 and P4 are the position points of two adjacent interpolation, and P5 is P0P n2 The intersection point with the arc, P6 is P3P4 and P0P n2 The intersection of P1P2 and P0P n2 The intersection of L1, L2, L3, L4, L5, L6 is the position segment P n1 P n2 , P n2 P n3 , P n3 P n4 , P n4 P n5 , P n5 P n6 , P n6 P n7The corresponding length; for the position micro-segment, the maximum speed of the straight line segment and the arc segment is v max , the maximum angular velocity is w max , the maximum acceleration is a max , maximum angular acceleration α max , set the maximum normal acceleration equal to the maximum acceleration;

[0039] For the position endpoint P n2 Corresponding to the transition arc segment, set P3P4 perpendicular to P0P n2 ; Set the bow height error generated at the highest speed to e ch , i.e. P5P6; set the contour error generated by interpolation at the highest speed to e co , that is, P n2 P6; Set line segment P n1 P n2 With line segment P n2 P n3 The angle between them is θ, that is, ∠P n1 P n2 P n3 ;

[0040] The position arc transition in the pose arc transition model, where the arc error radius r err The formula is expressed as:

[0041] r err / (e co -e ch +r err ) = sin(θ / 2);

[0042] r err =sin(θ / 2)×(e co -e ch ) / (1-sin(θ / 2));

[0043] When the transition arc radius is too long, select half of the smaller segment of the adjacent line segment as the transition arc restriction condition to meet the transition length requirement of the straight line segment and determine the arc radius r. len for:

[0044] r len =min(L1,L2) / (2tan(θ / 2));

[0045] According to the arc error radius and arc radius, the preset transition arc radius r is obtained. n =min(r err ,r len );

[0046] According to the normal acceleration of the arc and the preset transition arc radius, the allowable speed of the arc segment is:

[0047]

[0048] Then the velocity constraint of the arc segment is v n =min(v rmax ,v max ), the arc radius and speed constraints of the arc segments corresponding to other position endpoints can be obtained in the same way;

[0049] The calculation for the posture micro-segments is similar as above.

[0050] Further, step S6 is specifically as follows:

[0051] Use T-type velocity interpolation to obtain the planning time of the position and posture of the straight line segment and the arc segment, compare and analyze the posture planning time, determine whether the shorter time corresponds to the position segment or the posture segment, and determine the type corresponding to the time-lengthened segment, that is:

[0052] In the process of posture synchronization in each trajectory segment, the position planning time T is obtained according to the T-type speed planning. v and posture planning time T w , the following three situations may occur:

[0053] When T v =T w ,At this time, the position and attitude planning time are consistent, and no synchronous adjustment of the position and attitude is performed;

[0054] When T v >T w ,At this time, the position planning time is greater than the attitude planning time, and the attitude planning time is synchronously extended;

[0055] When T v <T w At this time, the position planning time is less than the attitude planning time, and the position planning time is synchronously lengthened.

[0056] Furthermore, step S7 specifically includes:

[0057] S71. Determine the T-type speed time synchronization method according to the posture synchronization requirements; obtain different posture synchronization methods by setting time limits; set the case where the terminal speed remains unchanged as the initial posture synchronization method, and set the case where the terminal speed changes as the changed posture synchronization method;

[0058] S72, adjust all the arc segments of the posture according to the maximum speed v that can be achieved by moving half of the arc path from 0 halfS As the arc start and end speed limit, the arc start and end speed limit is obtained as v = min (v n ,v halfS ), where v nThe arc segment speed constraint is used to ensure that the arc segment start and end speeds of the position and attitude meet the posture synchronization requirements no matter how they decrease, and the initial posture synchronization method described in step S71 is used for planning;

[0059] S73. For the posture synchronization of the straight segment, for the time period that needs to be extended, the initial posture synchronization method is first used for planning; when the initial posture synchronization method does not meet the synchronization time requirement, the modified posture synchronization method is used for planning; if the synchronization time requirement is still not met, the speed is adjusted according to the specific situation, that is:

[0060] If the currently planned straight line segment is not the first straight line segment in the forward-looking segment, the speed of the straight line segment is adjusted, and the speed is accelerated from 0 to the maximum speed along the halfway point of the straight line segment as the initial and final speeds at both ends of the current straight line; if the straight line segment is the first straight line segment in the forward-looking segment, the above adjustment method will reduce the initial speed, and the value of the previous forward-looking will be called as the current forward-looking speed.

[0061] More specifically, step S71 is as follows:

[0062] Set the speed curve K0 of acceleration-constant speed-deceleration to be elongated, and v s ≤v e , remember v s represents the starting point speed, v e Indicates the terminal speed, sets the planning boundaries B1, B2, B3, B4, establishes the speed planning curves K1, K2, K3, K4, and records the synchronization time as T F ; The T-type speed time synchronization method is specifically as follows:

[0063] The time limit T corresponding to B1 B1 for:

[0064]

[0065] Where S represents the length of the corresponding segment;

[0066] The time limit T corresponding to B2 B2 for:

[0067]

[0068] The time limit T corresponding to B3 B3 for:

[0069]

[0070] The time limit T corresponding to B4 B4 for:

[0071]

[0072] Where v e_down Indicates the final velocity of descent;

[0073] If the synchronization time satisfies T F <T B1 , then reduce the speed of the constant speed section, synchronize K0 with speed curve K1, that is, plan in the manner of acceleration-constant speed-deceleration; where v m Reduce to v m1 The acceleration, constant speed and deceleration time are T K1_1 , T K1_2 , T K1_3 , the formula is:

[0074]

[0075] If the synchronization time satisfies T B1 ≤T F <T B2 , then the speed curve K2 is synchronized with K0, that is, the planning is carried out in the manner of constant speed-acceleration-constant speed; the corresponding time of each section is T K2_1 , T K2_2 , T K2_3 , the formula is:

[0076]

[0077] If the synchronization time satisfies T B2 ≤T F <T B3 , then the speed curve K3 is synchronized with K0, that is, the planning is carried out in a constant speed-acceleration manner; where v e Reduce to v e-K3 , the corresponding time of each segment is T K3_1 , T K3_2 , the formula is:

[0078]

[0079] If the synchronization time satisfies T B3 ≤T F <T B4 , reduce the terminal speed, synchronize K0 with speed curve K4, that is, plan in a deceleration-constant speed manner. e Reduce to v e-K4 , the corresponding time of each segment is T K4_1 , T K4_2 , the formula is:

[0080]

[0081] Furthermore, step S8 specifically includes:

[0082] S81, for non-last forward-looking, output the other data in the forward-looking segment except the last arc segment and the two remaining straight lines, save the data of the last arc segment and the two straight lines, and then start the loop execution to perform the next forward-looking;

[0083] S82, for the last look-ahead, all data are output;

[0084] S83, planning the processing trajectory in real time according to the output data and the adjustment speed.

[0085] Based on the above technical solution, the present invention has the following beneficial effects:

[0086] 1. The present invention proposes a point cloud trajectory planning method suitable for curved surface components, constructs a robot grinding and processing system, and meets the grinding and processing needs of countless curved surface components.

[0087] 2. The present invention proposes a point cloud posture optimization method, which adopts and unifies the constraints of the robot's joint angle and maneuverability, takes the posture optimization of the surface machining process as the goal, establishes the optimal nonlinear objective function of the robot's machining posture, and uses a genetic algorithm to obtain the optimal machining posture.

[0088] 3. The present invention proposes a posture arc transition model, which comprehensively considers constraints such as contour error, bow height error, speed, acceleration, angular velocity, angular acceleration, posture change, etc., and provides posture synchronization initialization data.

[0089] 4. The present invention proposes a multi-boundary analytical posture synchronization method, which not only takes into account the speed and acceleration of the position, but also takes into account the speed and acceleration of the posture change. The analytical synchronization method is used to achieve posture synchronization and avoid sudden changes in posture.

[0090] 5. The present invention proposes a T-type speed and posture synchronization method, establishes time stretching strategies for different situations, and classifies them into an initial posture synchronization method and a changed posture synchronization method according to the final speed change. By reducing the constant speed or final speed of the section to be extended, the initial speed change is avoided and the processing efficiency is improved.

[0091] 6. The present invention proposes a nested posture synchronization look-ahead algorithm to strictly ensure the posture synchronization of the current segment and avoid the influence of the current posture look-ahead on the result of the previous posture look-ahead. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0093] Figure 1 It is the overall flow chart of the method proposed by the present invention;

[0094] Figure 2 Polishing processing systems for robots;

[0095] Figure 3 To obtain the local processing pose;

[0096] Figure 4 To optimize the local processing posture;

[0097] Figure 5 Flow chart of the posture synchronization forward planning algorithm in the method proposed by the present invention;

[0098] Figure 6 The pose transition model diagram in the method proposed by the present invention;

[0099] Figure 7 Schematic diagram of the posture synchronization method in the method proposed by the present invention;

[0100] Figure 8 For comparison method location map;

[0101] Fig. 9 To compare the method posture diagram;

[0102] Fig.10 Speed ​​graphs planned for comparative methods;

[0103] Fig.11 Angular velocity diagrams planned for comparison methods;

[0104] Fig.12 This is a location diagram of the method proposed by the present invention;

[0105] Fig.13 This is a posture diagram of the method proposed by the present invention;

[0106] Fig.14 A speed map planned for the method proposed in the present invention;

[0107] Fig.15 The angular velocity diagram planned by the method proposed in the present invention.

[0108] The reference numerals in the figure are: 1-industrial robot, 2-constant force flange, 3-eccentric grinder, 4-binocular vision, 5-curved composite component, 6-storage table. DETAILED DESCRIPTION

[0109] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0110] Although the steps in the present invention are arranged with numbers, they are not used to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a certain step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used in this article involves and covers any and all possible combinations of one or more of the associated listed items.

[0111] like Figure 1 As shown, the present invention proposes a point cloud trajectory planning method for a robot polishing system, which specifically includes the following steps:

[0112] S1. Establish the hardware configuration of the industrial robot grinding system, including an industrial robot, a constant force flange, an eccentric grinder, a binocular vision measurement device, a curved composite component and a storage table;

[0113] like Figure 2 As shown, the constant force flange is fixedly connected to the eccentric grinder and installed at the end of the robot, which is responsible for grinding and controlling the processing force to be stable at a constant value; the curved composite component is placed on the storage table; the binocular vision measurement device is placed on the ground at a certain distance from the processing area to ensure that the curved composite component, that is, the component to be processed, is in a better position in the field of view.

[0114] S2. Use a binocular vision measuring device to establish a coordinate system, and use a spherical scanner in the binocular vision measuring device to obtain the three-dimensional point cloud information corresponding to the component to be processed;

[0115] More specifically, in step S2, a robot base coordinate system, a workpiece coordinate system, a wrist coordinate system, and a tool coordinate system are established using a binocular vision measuring device; a spherical scanner in the binocular vision measuring device is used to obtain a point cloud of the component to be processed relative to the workpiece coordinate system; and then, based on the conversion relationship between the workpiece coordinate system and the robot base coordinate system, the three-dimensional point cloud information of the component to be processed under the robot base coordinate system is obtained.

[0116] S3, performing filtering sampling preprocessing on the three-dimensional point cloud information obtained in step S2, and obtaining the robot processing position and posture through a point cloud algorithm processing method;

[0117] As a preferred implementation, step S3 specifically includes:

[0118] S31. Use statistical filtering to remove noise data; perform statistics on the neighborhood of each point to obtain a result of approximate Gaussian distribution, and calculate the average distance d and standard deviation of each point to all adjacent points. σ , determine the standard deviation coefficient std, remove the points whose average distance is outside the standard range D = d + std × σ, that is, remove the points that are relatively "scattered" from the surrounding neighboring points;

[0119] S32, using the voxel downsampling method to obtain simplified point cloud data, determining rows, columns, and layers according to the maximum and minimum values ​​of the point cloud and the grid size, placing each point cloud data into the corresponding grid according to the coordinates, and calculating the centroid of the non-empty grid to replace all points in the voxel;

[0120] S33, using two planes separated by a distance d to intercept a point cloud layer, projecting the point cloud layer onto a middle plane between the two planes by projection, and obtaining initial path points; using a NURBS curve to fit the path points, and obtaining discrete processing positions with equidistant steps;

[0121] S34, using a point cloud normal vector algorithm, find a plane of point cloud fitting within a sphere of radius R of the processing position obtained in step S33, select any key point on one side of the curved surface, and obtain a normal vector of the fitting plane toward the key point as the normal vector of the processing position;

[0122] The rational expression of the k-order NURBS curve given in this embodiment is as follows:

[0123]

[0124] Among them, ω i represents the control factor; d i Represents the control vertex; N i,k (u) represents the k-order canonical B-spline basis function determined by the DeBoer-Cox recursion formula.

[0125] S35. Use the normal vector obtained in step S34 as the z-axis, and set the x and y axes arbitrarily while satisfying the xyz right-hand coordinate system principle to calculate the Euler angle of this xyz coordinate system relative to the robot base coordinate system; use the Euler angle and the coordinates of the processing position obtained in S33 as the processing posture; the robot base coordinate system is established by a binocular vision measurement device.

[0126] S4, according to the processing posture obtained in step S3, using the robot processing posture constraint conditions, optimizing the robot processing posture, and obtaining the final robot posture optimization result;

[0127] As a preferred implementation, step S4 specifically includes:

[0128] S41. Calculate the joint angles corresponding to different robot terminal positions based on the robot inverse kinematics, obtain the joint range based on the robot technical parameters, and take the ratio of the joint position to the median of the joint range to the joint motion range as the robot joint index R. θ , as follows:

[0129]

[0130] in, n represents the number of robot joints, θ imin Indicates the minimum limit of the joint corresponding to joint i, 1≤i≤6, θ imax represents the maximum limit of the joint corresponding to joint i, θ imid Indicates that joint i corresponds to the middle limit of the joint, and θ i represents the angle of each joint, and θ imin ≤θ i ≤θ imax ;

[0131] S42. Maneuverability is a comprehensive measure of the robot's ability to move in all directions in a certain posture. The reciprocal of maneuverability is used as the robot's dexterity index R ω , the formula is:

[0132]

[0133] in, is the operability, J(θ) is the Jacobian matrix;

[0134] S43. On the basis of step S41 and step S42, dimensionless variables of robot joint index and robot dexterity index are obtained, and the robot end posture change angle corresponding to the processing path point is used as a variable, and a nonlinear objective function for obtaining the optimal value of the robot processing posture is established. A genetic algorithm is used to optimize and solve the target to obtain the optimal posture change angle; the obtained posture is used as the final robot processing posture.

[0135] S5. According to the posture result obtained in step S4, select N posture forward trajectory segments, N ≥ 3, and perform initialization parameter setting according to the posture transition model and speed constraints;

[0136] As a preferred implementation, step S5 in this embodiment specifically includes:

[0137] S51. Perform position transition initialization. For the first look-ahead initialization, obtain the arc radius under the maximum error, obtain half of the trajectory segment as the arc radius of the arc transition, take the smaller of the two radii as the initialization value of the arc radius, and set the initial speed of the first look-ahead segment and the terminal speed of the last segment to 0. The arc segment speed constraint is set according to the comprehensive restrictions of acceleration, normal acceleration, and maximum speed. The speed and acceleration constraints of the straight segment are respectively set according to the maximum speed and maximum acceleration limits. The connection point constraint conditions of the straight segment and the arc segment are based on the arc segment speed and acceleration constraints.

[0138] In this embodiment, Figure 6 As shown, note P n1 , P n2 , P n3 , P n4 , P n5 , P n6 , P n7 is the endpoint of the position micro segment, P0 is the endpoint of the position segment P n2 Corresponding to the center point of the transition arc segment, P1 and P2 are the tangent points of the position transition arc segment and the position straight line segment, P3 and P4 are the position points of two adjacent interpolation, and P5 is P0P n2 The intersection point with the arc, P6 is P3P4 and P0P n2 The intersection of P1P2 and P0P n2 The intersection of L1, L2, L3, L4, L5, L6 is the position segment P n1 P n2 , P n2 P n3 , P n3 P n4 , P n4 P n5 , P n5 P n6 , P n6 P n7 The corresponding length; Z n1 , Z n2 , Z n3 , Z n4 , Z n5 , Z n6 , Z n7 is the endpoint of the posture micro-segment, Z0 is the endpoint of the posture segment Z n2 Corresponding to the center point of the transition arc segment, Z1 and Z2 are the tangent points of the posture transition arc segment and the posture straight line segment, Z3 and Z4 are the posture points of two adjacent interpolations, and Z5 is Z0Z n2 The intersection point with the arc, Z6 is Z3Z4 and Z0Z n2 The intersection of Z1Z2 and Z0Z n2The intersection of is the intersection of and , A1, A2, A1, A1, A1, A1 represent the posture segment Z n1 Z n2 , Z n2 Z n3 , Z n3 Z n4 , Z n4 Z n5 , Z n5 Z n6 , Z n6 Z n7 The corresponding length.

[0139] For the position micro-segment, the maximum speed of the straight line segment and the arc segment is v max , the maximum angular velocity is w max , the maximum acceleration is a max , maximum angular acceleration α max , set the maximum normal acceleration equal to the maximum acceleration. n2 Corresponding to the transition arc segment, set P3P4 perpendicular to P0P n2 ; Set the bow height error generated at the highest speed to e ch , that is, the contour error generated by interpolation at the highest speed set by P5P6 is e co , that is, P n2 P6; Set line segment P n1 P n2 With line segment P n2 P n3 The angle between them is θ, that is, ∠P n1 P n2 P n3 .

[0140] Depend on Figure 6 Position arc transition in the pose arc transition model, where the arc error radius r err The formula is expressed as:

[0141] r err / (e co -e ch +r err ) = sin(θ / 2);

[0142] r err =sin(θ / 2)×(e co -e ch ) / (1-sin(θ / 2));

[0143] When the transition arc radius is too long, select half of the smaller segment of the adjacent line segment as the transition arc restriction condition to meet the transition length requirement of the straight line segment and determine the arc radius r. len for:

[0144] r len =min(L1,L2) / (2tan(θ / 2));

[0145] According to the arc error radius and arc radius, the preset transition arc radius r is obtained. n =min(r err ,r len );

[0146] According to the normal acceleration of the arc and the preset transition arc radius, the allowable speed of the arc segment is:

[0147]

[0148] Then the velocity constraint of the arc segment is v n =min(v rmax ,v max ), the arc radius and speed constraints of the arc segments corresponding to other position endpoints can be obtained in the same way;

[0149] The calculation for the posture micro-segments is similar as above.

[0150] S52, for initialization other than the first look-ahead, the look-ahead starting speed is set according to the final speed outputted by the last look-ahead, and the preset manner of the remaining variables is the same as the first look-ahead setting in step S51;

[0151] S53, on the basis of steps S51 and S52, the boundary speed is judged as acceleration / deceleration type according to the boundary speeds at both ends of the straight line, and a boundary speed determination method is formulated; during the acceleration process, the end speed constraint is reduced to the maximum speed to which the starting speed of the straight line segment can be accelerated; during the deceleration process, the starting speed constraint is reduced to the maximum speed to which the end speed of the straight line segment can be reversely accelerated, and the parameter setting of the position transition initialization is updated with this speed constraint;

[0152] S54. Similarly, perform posture transition initialization according to the process described in steps S51-S53.

[0153] S6, obtaining the planning time of the straight line segment and arc segment of the robot position and posture through the T-type speed interpolation method, and performing position and posture synchronization judgment to determine the information of the lengthened segment to be synchronized;

[0154] As a preferred implementation, step S6 is specifically as follows:

[0155] Use T-type velocity interpolation to obtain the planning time of the position and posture of the straight line segment and the arc segment, compare and analyze the posture planning time, determine whether the shorter time corresponds to the position segment or the posture segment, and determine the type corresponding to the time-lengthened segment, that is:

[0156] In the process of posture synchronization in each trajectory segment, the position planning time T is obtained according to the T-type speed planning. v and posture planning time T w , the following three situations may occur:

[0157] When T v =T w ,At this time, the position and attitude planning time are consistent, and no synchronous adjustment of the position and attitude is performed;

[0158] When T v >T w ,At this time, the position planning time is greater than the attitude planning time, and the attitude planning time is synchronously extended;

[0159] When T v <T w At this time, the position planning time is less than the attitude planning time, and the position planning time is synchronously lengthened.

[0160] S7, based on the posture synchronization judgment result of step S6, adjusting the posture synchronization of the arc segment and the straight line segment respectively by a speed synchronization method;

[0161] As a preferred implementation, step S7 specifically includes:

[0162] S71. Determine the T-type speed time synchronization method according to the posture synchronization requirements; obtain different posture synchronization methods by setting time limits; set the case where the terminal speed remains unchanged as the initial posture synchronization method, and set the case where the terminal speed changes as the changed posture synchronization method;

[0163] like Figure 7 As shown, the T-type speed time synchronization method in this embodiment is specifically as follows:

[0164] Set the speed curve K0 of acceleration-constant speed-deceleration to be elongated, and v s ≤v e , remember v s represents the starting point speed, v e Indicates the terminal speed, sets the planning boundaries B1, B2, B3, B4, establishes the speed planning curves K1, K2, K3, K4, and records the synchronization time as T F ; The T-type speed time synchronization method is specifically as follows:

[0165] The time limit T corresponding to B1 B1 for:

[0166]

[0167] Where S represents the length of the corresponding segment;

[0168] The time limit T corresponding to B2B2 for:

[0169]

[0170] The time limit T corresponding to B3 B3 for:

[0171]

[0172] The time limit T corresponding to B4 B4 for:

[0173]

[0174] Where v e_down Indicates the final velocity of descent;

[0175] If the synchronization time satisfies T F <T B1 , then reduce the speed of the constant speed section, synchronize K0 with speed curve K1, that is, plan in the manner of acceleration-constant speed-deceleration; where v m Reduce to v m1 The acceleration, constant speed and deceleration time are T K1_1 , T K1_2 , T K1_3 , the formula is:

[0176]

[0177] If the synchronization time satisfies T B1 ≤T F <T B2 , then the speed curve K2 is synchronized with K0, that is, the planning is carried out in the manner of constant speed-acceleration-constant speed; the corresponding time of each section is T K2_1 , T K2_2 , T K2_3 , the formula is:

[0178]

[0179] If the synchronization time satisfies T B2 ≤T F <T B3 , then the speed curve K3 is synchronized with K0, that is, the planning is carried out in a constant speed-acceleration manner; where v e Reduce to v e-K3 , the corresponding time of each segment is T K3_1 , T K3_2 , the formula is:

[0180]

[0181] If the synchronization time satisfies T B3 ≤TF <T B4 , reduce the terminal speed, synchronize K0 with speed curve K4, that is, plan in a deceleration-constant speed manner. e Reduce to v e-K4 , the corresponding time of each segment is T K4_1 , T K4_2 , the formula is:

[0182]

[0183] S72, adjust all the arc segments of the posture according to the maximum speed v that can be achieved by moving half of the arc path from 0 halfS As the arc start and end speed limit, the arc start and end speed limit is obtained as v = min (v n ,v halfS ), where v n The arc segment speed constraint is used to ensure that the arc segment start and end speeds of the position and attitude meet the posture synchronization requirements no matter how they decrease, and the initial posture synchronization method described in step S71 is used for planning;

[0184] S73. For the posture synchronization of the straight segment, for the time period that needs to be extended, the initial posture synchronization method is first used for planning; when the initial posture synchronization method does not meet the synchronization time requirement, the modified posture synchronization method is used for planning; if the synchronization time requirement is still not met, the speed is adjusted according to the specific situation, that is:

[0185] If the currently planned straight line segment is not the first straight line segment in the forward-looking segment, the speed of the straight line segment is adjusted, and the speed is accelerated from 0 to the maximum speed along the halfway point of the straight line segment as the initial and final speeds at both ends of the current straight line; if the straight line segment is the first straight line segment in the forward-looking segment, the above adjustment method will reduce the initial speed, and the value of the previous forward-looking will be called as the current forward-looking speed.

[0186] S8, and determine whether the last posture is synchronously forward-looking, and make different data output and storage decisions. The output data plans the processing trajectory in real time according to the adjustment speed.

[0187] As a preferred implementation, step S8 specifically includes:

[0188] S81, for non-last forward-looking, output the other data in the forward-looking segment except the last arc segment and the two remaining straight lines, save the data of the last arc segment and the two straight lines, and then start the loop execution to perform the next forward-looking;

[0189] S82, for the last look-ahead, all data are output;

[0190] S83, planning the processing trajectory in real time according to the output data and the adjustment speed.

[0191] In this embodiment, the above method can also be summarized as follows: Figure 5 The pose synchronization forward planning algorithm flow is shown.

[0192] The following experimental cases are given to verify the effect of the method proposed in the present invention:

[0193] Verification 1: A part of the point cloud image is selected as the simulated polishing area, and the trajectory directly obtained by the point cloud slicing algorithm and the normal vector fitting algorithm has many problems such as redundant position points and inconsistent normal vectors. Figure 3 After path fitting, normal alignment, and posture optimization, the trajectory becomes smoother and simpler, such as Figure 4 .

[0194] Verification 2: Set the posture synchronization look-ahead algorithm to the posture shown in Table 1 for planning verification. The specific processing parameters are set as follows: maximum speed 10 mm / s, maximum acceleration 100 mm / s 2 , maximum contour error 0.05mm, maximum bow height error 0.01mm; maximum angular velocity 10rad / s, maximum angular acceleration 100rad / s 2 , the maximum angle profile error is 0.05rad, and the maximum angle bow height error is 0.01rad. Figure 8-11 For the processing results using the comparison method, Figure 12-15 This is the processing result of the method proposed in the present invention. It can be seen that the planned path and posture of the present invention are smoother, and the planned speed and angular velocity are better. To plan the same path, the comparison method takes 1936ms, while the method of the present invention takes 1678ms, which improves the efficiency by 13.33%.

[0195] Table 1 Point data of the posture to be planned

[0196]

[0197] In summary, the present invention obtains and optimizes the processing posture according to the real-time point cloud to realize the synchronous forward planning of posture; solves the error problem of the digital model generation trajectory caused by the inability to position the curved composite component, avoids the posture mutation caused by the small path-large posture change, meets the curved surface grinding processing requirements, and improves the processing efficiency.

[0198] It should be noted that it is obvious to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be regarded as exemplary and non-restrictive from any point of view, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any reference numerals in the claims should not be regarded as limiting the claims involved.

[0199] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A point cloud trajectory planning method for a robot polishing system, characterized in that: include: S1. Establish the hardware configuration of the industrial robot grinding system, including an industrial robot, a constant force flange, an eccentric grinder, a binocular vision measuring device, a curved composite component and a storage table; the constant force flange is fixedly connected to the eccentric grinder and installed at the end of the robot, responsible for grinding and controlling the processing force to be stable at a constant value; the curved composite component is placed on the storage table; the binocular vision measuring device is placed on the ground at a certain distance from the processing area to ensure that the curved composite component, that is, the component to be processed, is in a better position in the field of view; S2. Use a binocular vision measuring device to establish a coordinate system, and use a spherical scanner in the binocular vision measuring device to obtain the three-dimensional point cloud information corresponding to the component to be processed; S3, performing filtering sampling preprocessing on the three-dimensional point cloud information obtained in step S2, and obtaining the robot processing position and posture through a point cloud algorithm processing method; S4, according to the processing posture obtained in step S3, using the robot processing posture constraint conditions, optimizing the robot processing posture, and obtaining the final robot posture optimization result; S5. According to the posture result obtained in step S4, select N posture forward trajectory segments, N ≥ 3, and perform initialization parameter setting according to the posture transition model and speed constraints; S6, obtaining the planning time of the straight line segment and arc segment of the robot position and posture through the T-type speed interpolation method, and performing the position and posture synchronization judgment result to determine the information of the lengthened segment to be synchronized; S7, based on the posture synchronization judgment result of step S6, adjusting the posture synchronization of the arc segment and the straight line segment respectively by a speed synchronization method; S8, and determine whether the last posture is synchronized forward, and make different data output and storage decisions; the output data plans the processing trajectory in real time according to the adjustment speed.

2. A point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S2 is specifically as follows: The robot base coordinate system, workpiece coordinate system, wrist coordinate system and tool coordinate system are established by using binocular vision measuring equipment. The ball scan in the binocular vision measuring equipment is used to obtain the point cloud of the component to be processed relative to the workpiece coordinate system. Then, based on the conversion relationship between the workpiece coordinate system and the robot base coordinate system, the three-dimensional point cloud information of the component to be processed under the robot base coordinate system is obtained.

3. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S3 specifically includes: S31. Use statistical filtering to remove noise data; perform statistics on the neighborhood of each point to obtain a result of approximate Gaussian distribution, and calculate the average distance d and standard deviation of each point to all adjacent points. σ , determine the standard deviation coefficient std, remove the points whose average distance is outside the standard range D = d + std × σ, that is, remove the points whose surrounding neighboring points are relatively "scattered"; S32, using the voxel downsampling method to obtain simplified point cloud data, determining rows, columns, and layers according to the maximum and minimum values ​​of the point cloud and the grid size, placing each point cloud data into the corresponding grid according to the coordinates, and calculating the centroid of the non-empty grid to replace all points in the voxel; S33, using two planes separated by a distance d to intercept a point cloud layer, projecting the point cloud layer onto a middle plane between the two planes by projection, and obtaining initial path points; using a NURBS curve to fit the path points, and obtaining discrete processing positions with equidistant steps; S34, using a point cloud normal vector algorithm, find a plane of point cloud fitting within a sphere of radius R of the processing position obtained in step S33, select any key point on one side of the curved surface, and obtain a normal vector of the fitting plane toward the key point as the normal vector of the processing position; S35. Use the normal vector obtained in step S34 as the z-axis, and set the x and y axes arbitrarily while satisfying the xyz right-hand coordinate system principle to calculate the Euler angle of this xyz coordinate system relative to the robot base coordinate system; use the Euler angle and the coordinates of the processing position obtained in S33 as the processing posture; the robot base coordinate system is established by a binocular vision measurement device.

4. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S4 specifically includes: S41. Calculate the joint angles corresponding to different robot terminal positions based on the robot inverse kinematics, obtain the joint range based on the robot technical parameters, and take the ratio of the joint position to the median of the joint range to the joint motion range as the robot joint index R. θ , as follows: in, n represents the number of robot joints, θ imin Indicates the minimum limit of the joint corresponding to joint i, 1≤i≤6, θ imax represents the maximum limit of the joint corresponding to joint i, θ imid Indicates that joint i corresponds to the middle limit of the joint, and θ i represents the angle of each joint, and θ imin ≤θ i ≤θ imax ; S42. Use maneuverability to comprehensively measure the robot's ability to move in all directions in a certain posture, and take the reciprocal of the maneuverability as the robot's dexterity index R. ω , the formula is: in, is the operability, J(θ) is the Jacobian matrix; S43. On the basis of step S41 and step S42, dimensionless variables of robot joint index and robot dexterity index are obtained, and the robot end posture change angle corresponding to the processing path point is used as a variable, and a nonlinear objective function for obtaining the optimal value of the robot processing posture is established. A genetic algorithm is used to optimize and solve the target to obtain the optimal posture change angle; the obtained posture is used as the final robot processing posture.

5. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S5 specifically includes: S51, perform position transition initialization; for the first forward initialization, obtain the arc radius under the maximum error, obtain half of the trajectory segment as the arc radius of the arc transition, take the smaller of the two radii as the initial value of the arc radius, and set the initial speed of the first forward segment and the terminal speed of the last segment to 0; the arc segment speed constraint is set according to the comprehensive restrictions of acceleration, normal acceleration, and maximum speed; the speed and acceleration constraints of the straight segment are respectively set according to the maximum speed and maximum acceleration restrictions; the connection point constraint conditions of the straight segment and the arc segment are set according to the arc segment speed and acceleration constraints; S52, for initialization other than the first look-ahead, the look-ahead starting speed is set according to the final speed outputted by the last look-ahead, and the preset manner of the remaining variables is the same as the first look-ahead setting in step S51; S53, on the basis of steps S51 and S52, the boundary speed is judged as acceleration / deceleration type according to the boundary speeds at both ends of the straight line, and a boundary speed determination method is formulated; during the acceleration process, the end speed constraint is reduced to the maximum speed to which the starting speed of the straight line segment can be accelerated; during the deceleration process, the starting speed constraint is reduced to the maximum speed to which the end speed of the straight line segment can be reversely accelerated, and the parameter setting of the position transition initialization is updated with this speed constraint; S54. Similarly, perform posture transition initialization according to the process described in steps S51-S53.

6. A point cloud trajectory planning method for a robot polishing system according to claim 5, characterized in that: Step S51 is specifically as follows: Note: n1 , P n2 , P n3 , P n4 , P n5 , P n6 , P n7 is the endpoint of the position micro segment, P0 is the endpoint of the position segment P n2 Corresponding to the center point of the transition arc segment, P1 and P2 are the tangent points of the position transition arc segment and the position straight line segment, P3 and P4 are the position points of two adjacent interpolation, and P5 is P0P n2 The intersection point with the arc, P6 is P3P4 and P0P n2 The intersection of P1P2 and P0P n2 The intersection of L1, L2, L3, L4, L5, L6 is the position segment P n1 P n2 , P n2 P n3 , P n3 P n4 , P n4 P n5 , P n5 P n6 , P n6 P n7 The corresponding length; for the position micro-segment, the maximum speed of the straight line segment and the arc segment is v max , the maximum angular velocity is w max , the maximum acceleration is a max , maximum angular acceleration α max , set the maximum normal acceleration equal to the maximum acceleration; For the position endpoint P n2 Corresponding to the transition arc segment, set P3P4 perpendicular to P0P n2 ; Set the bow height error generated at the highest speed to e ch , i.e. P5P6; set the contour error generated by interpolation at the highest speed to e co , that is, P n2 P6; Set line segment P n1 P n2 With line segment P n2 P n3 The angle between them is θ, that is, ∠P n1 P n2 P n3 ; The position arc transition in the pose arc transition model, where the arc error radius r err The formula is expressed as: r err / (e co -e ch +r err )=sin(θ / 2); r err =sin(θ / 2)×(e co -e ch ) / (1-sin(θ / 2)); When the transition arc radius is too long, select half of the smaller segment of the adjacent line segment as the transition arc restriction condition to meet the transition length requirement of the straight line segment and determine the arc radius r. len for: <h2 style=";text-align:left;direction:ltr">r<h2 style=";text-align:left;direction:ltr"> len <h2 style=";text-align:left;direction:ltr"> (min(L1,L2) / (2tan(θ / 2))) According to the arc error radius and arc radius, the preset transition arc radius r is obtained. n =min(r err ,r len ); According to the normal acceleration of the arc and the preset transition arc radius, the allowable speed of the arc segment is: Then the velocity constraint of the arc segment is v n =min(v rmax ,v max ), the arc radius and speed constraints of the arc segments corresponding to other position endpoints can be obtained in the same way; The calculation for the posture micro-segments is similar as above.

7. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S6 is specifically as follows: Use T-type velocity interpolation to obtain the planning time of the position and posture of the straight line segment and the arc segment, compare and analyze the posture planning time, determine whether the shorter time corresponds to the position segment or the posture segment, and determine the type corresponding to the time-lengthened segment, that is: In the process of posture synchronization in each trajectory segment, the position planning time T is obtained according to the T-type speed planning. v and posture planning time T w , the following three situations may occur: When T v =T w ,At this time, the position and attitude planning time are consistent, and no synchronous adjustment of the position and attitude is performed; When T v >T w ,At this time, the position planning time is greater than the attitude planning time, and the attitude planning time is synchronously extended; When T v <T w At this time, the position planning time is less than the attitude planning time, and the position planning time is synchronously lengthened.

8. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S7 is specifically as follows: S71. Determine the T-type speed time synchronization method according to the posture synchronization requirements; obtain different posture synchronization methods by setting time limits; set the case where the terminal speed remains unchanged as the initial posture synchronization method, and set the case where the terminal speed changes as the changed posture synchronization method; S72, adjust all the arc segments of the posture according to the maximum speed v that can be achieved by moving half of the arc path from 0 halfS As the arc start and end speed limit, the arc start and end speed limit is obtained as v = min (v n ,v halfS ), where v n The arc segment speed constraint is used to ensure that the arc segment start and end speeds of the position and attitude meet the posture synchronization requirements no matter how they decrease, and the initial posture synchronization method described in step S71 is used for planning; S73. For the posture synchronization of the straight segment, for the time period that needs to be extended, the initial posture synchronization method is first used for planning; when the initial posture synchronization method does not meet the synchronization time requirement, the modified posture synchronization method is used for planning; if the synchronization time requirement is still not met, the speed is adjusted according to the specific situation, that is: If the currently planned straight line segment is not the first straight line segment in the forward-looking segment, the speed of the straight line segment is adjusted, and the speed is accelerated from 0 to the maximum speed along the halfway point of the straight line segment as the initial and final speeds at both ends of the current straight line; if the straight line segment is the first straight line segment in the forward-looking segment, the above adjustment method will reduce the initial speed, and the value of the previous forward-looking will be called as the current forward-looking speed.

9. The point cloud trajectory planning method for a robot polishing system according to claim 8, characterized in that: Step S71 specifically includes: Set the speed curve K0 of acceleration-constant speed-deceleration to be elongated, and v s ≤v e , remember v s represents the starting point speed, v e Indicates the terminal speed, and sets the planning boundaries B1, B2, B3, B4, establishes the speed planning curves K1, K2, K3, K4, and records the synchronization time as T F ; The T-type speed time synchronization method is specifically as follows: The time limit T corresponding to B1 B1 for: Where S represents the length of the corresponding segment; The time limit T corresponding to B2 B2 for: The time limit T corresponding to B3 B3 for: The time limit T corresponding to B4 B4 for: Where v e_down Indicates the final velocity of descent; If the synchronization time satisfies T F <T B1 , then reduce the speed of the constant speed section, synchronize K0 with speed curve K1, that is, plan in the manner of acceleration-constant speed-deceleration; where v m Reduce to v m1 The acceleration, constant speed and deceleration time are T K1_1 , T K1_2 , T K1_3 , the formula is: If the synchronization time satisfies T B1 ≤T F <T B2 , then the speed curve K2 is synchronized with K0, that is, the planning is carried out in the manner of constant speed-acceleration-constant speed; the corresponding time of each section is T K2_1 , T K2_2 , T K2_3 , the formula is: If the synchronization time satisfies T B2 ≤T F <T B3 , then the speed curve K3 is synchronized with K0, that is, the planning is carried out in a constant speed-acceleration manner; where v e Reduce to v e-K3 , the corresponding time of each segment is T K3_1 , T K3_2 , the formula is: If the synchronization time satisfies T B3 ≤T F <T B4 , reduce the terminal speed, synchronize K0 with speed curve K4, that is, plan in a deceleration-constant speed manner; where v e Reduce to v e-K4 , the corresponding time of each segment is T K4_1 , T K4_2 , the formula is:

10. The point cloud trajectory planning method for a robot polishing system according to claim 1, characterized in that: Step S8 specifically includes: S81, for non-last forward-looking, output the other data in the forward-looking segment except the last arc segment and the two remaining straight lines, save the data of the last arc segment and the two straight lines, and then start the loop execution to perform the next forward-looking; S82, for the last look-ahead, output all data; S83, planning the processing trajectory in real time according to the output data and the adjustment speed.

Images