Robot collaborative sewing method and system based on speed look-ahead control

Abstract

The invention provides a robot collaborative sewing method and system based on speed look-ahead control. The system comprises an image detection module, a pressure detection module, sewing equipment, a controller and a mechanical arm. The image detection module acquires image information of the fabric in real time and transmits the image information to the controller; the controller processes the image information to generate the position of a pressing point at the tail end and a motion path, introduces speed look-ahead control, optimizes a speed extreme value according to the motion path, completes speed planning of the whole track, and interpolates the planned track to obtain an interpolated result; the controller sends expected force to the pressure detection module, and the pressure detection module outputs the pressing height; the mechanical arm presses the height to execute the action to the fabric based on the interpolation result; the controller sends a control signal to the sewing equipment, and the sewing equipment carries out cloth feeding and sewing; the speed of the sewing equipment is matched with that of the mechanical arm, and cooperative sewing is carried out.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot collaborative sewing, and in particular relates to a robot collaborative sewing method and system based on speed forward control. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In the industrialized production of clothing, the emergence of various types of sewing machines has greatly reduced the requirements for the professional skills of operators, but the sewing process still relies on manual control of the position of the fabric and its coordination with the speed of the sewing machine.

[0004] How to use industrial robots to replace manual labor to complete fabric operations has become an urgent problem to be solved. Since fabric materials are anisotropic and non-uniform, their ability to resist deformation is low, which leads to wrinkles and deformation during the sewing process.

[0005] Most existing robotic fabric manipulation methods are based on geometric modeling, which detects and controls tension to compensate for deformation by establishing a physical model of the fabric. However, these model-based or force-controlled methods are difficult to accurately simulate the physical properties of fabrics, and the compensation effect is limited.

[0006] After searching, it was found that there are also existing solutions for sewing with speed control. Most of the existing methods plan fixed paths and speeds, and no real-time adjustments are made according to the actual situation. Summary of the invention

[0007] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a robot collaborative sewing method based on speed look-ahead control. The method performs path planning before sewing, adds a look-ahead speed correction process in the subsequent movement process, matches the sewing speed of each section, and thus avoids the generation of wrinkles in advance without having to strictly require fabric properties, etc.

[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0009] In a first aspect, a robot collaborative sewing system based on speed look-ahead control is disclosed, comprising: an image detection module, a pressure detection module, a sewing device, a controller and a robotic arm;

[0010] The image detection module collects image information of the fabric in real time and transmits it to the controller;

[0011] The pressure detection module obtains the pressing height of the end pressing the fabric in real time during the sewing process and feeds it back to the controller;

[0012] After processing the image information, the controller generates the pressing point position and motion path of the end, introduces speed look-ahead control, optimizes the speed extreme value according to the motion path, completes the speed planning of the entire trajectory, interpolates the planned trajectory, and obtains the interpolated result;

[0013] The controller then determines the position and posture of the end of the robot arm according to the interpolated result and the pressing height;

[0014] The robot arm performs actions to the fabric based on the end position;

[0015] The controller sends a control signal to the sewing device, and the sewing device performs cloth feeding and sewing;

[0016] The sewing device matches the speed of the robot arm to perform coordinated sewing.

[0017] As a further technical solution, the image detection module is a depth camera, and the image information of the fabric collected in real time is a color image and a depth image of the sewing environment.

[0018] In a second aspect, a robot collaborative sewing method based on speed forward control is disclosed, comprising:

[0019] Collecting image information of fabric;

[0020] Obtain the pressing height of the fabric at the end during sewing in real time;

[0021] After processing the image information, the pressing point position and motion path of the end are generated, and the speed look-ahead control is introduced. According to the motion path, the speed extreme value is optimized to complete the speed planning of the entire trajectory, and the planned trajectory is interpolated to obtain the interpolated result;

[0022] Determine the end position of the robot arm according to the interpolated result and the pressing height;

[0023] The robot arm performs actions to the fabric based on the end pose;

[0024] Sewing equipment is used for feeding and sewing;

[0025] The sewing equipment matches the speed of the mechanical arm to perform coordinated sewing.

[0026] As a further technical solution, the image information is processed, specifically including:

[0027] Extract image contours from image information and perform rough interpolation to obtain a sequence of contour points;

[0028] Based on the calibration, the pixel coordinates are converted to the robot coordinate system through matrix operation, and the pressing point position and motion path of the end of the robot are generated according to the contour points.

[0029] As a further technical solution, the image contour is extracted from the image information and a rough interpolation is performed to obtain a contour point sequence. The specific process is as follows:

[0030] The image information includes a color image and a depth image of the sewing environment;

[0031] Generate 3D point cloud images using depth images and save them; read RGB images, depth images and point cloud images;

[0032] Define the HSV range according to the color of the cloth, filter the image, erode and dilate the image, and use the Canny operator to detect the edge of the cloth;

[0033] Sampling the detected contours to form a dense point set;

[0034] Fitting NURBS curve according to sampling points;

[0035] Select a suitable step length Δu, starting from u=0, and calculate the corresponding C(u) point every other step length Δu to obtain a new interpolation point sequence;

[0036] The interpolated point set is used to generate a point cloud image to form the 3D outline of the cloth.

[0037] As a further technical solution, the specific process of generating the pressing point coordinates and the terminal path is as follows:

[0038] The interpolated coordinate points are converted from the camera coordinate system to the robot arm coordinate system as a point set through matrix operations;

[0039] Identify the x coordinates of the point set max 、x min and max ,y min , take the average value as the pressing point;

[0040] For adjacent cloth contour points P i and P i+1 , calculate the direction angle, obtain the rotation angle, and then calculate the end P end To contour point P i The translation vector of

[0041] Construct a transformation matrix to transform the contour point coordinates into the end path.

[0042] As a further technical solution, the speed forward control process is specifically as follows:

[0043] Constrain velocity and acceleration: set a minimum velocity v min and minimum acceleration a min , set the maximum speed v max and the maximum acceleration a max;

[0044] According to the obtained point set, calculate the curvature value k of each path point i Using the discrete curvature formula, for each point P on the path i , through its front and rear points P i-1 and P i+1 Calculate the coordinates of

[0045] Define a sliding window, calculate the curvature values ​​of n consecutive points in the path, and take the average rate of change to detect mutations;

[0046] Set the threshold k threshold ,like It is believed that P i+n-1 Speed ​​mutation point.

[0047] According to the above method, the adjacent velocity mutation point P is obtained. j and P k , calculate the path length between mutation points:

[0048] To determine the number of forward-looking segments, first calculate the average distance between two adjacent interpolation points:

[0049] Then calculate the number of forward-looking segments based on the above average value to ensure that enough points are planned;

[0050] When the speed changes greatly before and after the mutation point and the path is short, the acceleration requirement may not be met and the speed should be corrected;

[0051] Use NURBS curves to interpolate feed speed and position information in real time.

[0052] As a further technical solution, when using NURBS curves to interpolate feed speed and position information in real time, more points are inserted in the curve segment to ensure a smooth transition; and the number of points in the straight line segment is reduced to keep the path simple.

[0053] As a further technical solution, the speed is corrected, specifically:

[0054] If in the acceleration stage, from v i Accelerate to v i+1 , need to correct v i+1 , so that v′ i+1 ≤v i+1 ;

[0055] If it is in the deceleration stage, from v i Decelerate to v i+1 , need to be repaired i , so that v′ i ≤v i .

[0056] As a further technical solution, when processing short planning elements, before planning the long segment i, determine whether the i+1th segment is a short segment. If so, continue to index backward until the i+k+1th segment is found to be a long segment. At this time, the feed speed is adjusted in reverse from the i+kth segment to the i+1th segment, and then the terminal speed of the i-th curve segment is forward planned, and the planning of the larger parts of each curvature is completed in turn.

[0057] One or more of the above technical solutions have the following beneficial effects:

[0058] Based on the speed forward control, the sewing trajectory is planned, and a solution to improve the stability and fluency of the robot sewing operation is proposed. Through the NURBS curve fitting path, interpolation points can be flexibly generated as needed, and secondary interpolation can be performed to improve the trajectory accuracy and ensure stable speed. The feedback of visual information and force information is used to improve the accuracy of sewing operations, ensure the stability of path following, and reduce deviations in fabric movement. The technical solution of the present invention calculates the curvature of the path, and the speed setting is different at different curvatures. For errors that may exist during actual operation, the speed correction method mentioned is used to forward control the speed and make certain adjustments in advance. According to the speed of the robot arm, a signal is also sent to the sewing machine to dynamically control the speed of the sewing machine. At the same time, feedback is used to control the end pressing force in real time.

[0059] The technical solution of the present invention dynamically adjusts the target speed, solves the problem of coordinated control of the robot arm and the sewing machine, avoids the influence of wrinkles in advance, does not set strict requirements on the characteristics of the fabric, is applicable to fabrics of different materials, and has high versatility.

[0060] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0062] Figure 1 A robot sewing system according to an embodiment of the present invention;

[0063] Figure 2 This is a schematic diagram of a robot sewing process according to an embodiment of the present invention;

[0064] Figure 3 This is a flow chart of speed look-ahead control according to an embodiment of the present invention;

[0065] Figure 4 This is a schematic diagram of a deceleration mutation point according to an embodiment of the present invention;

[0066] Figure 5 Schematic diagram of the velocity replanning process of the short NURBS planner according to an embodiment of the present invention. DETAILED DESCRIPTION

[0067] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0068] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0069] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0070] In order to improve the coordination between the robot arm and the sewing machine in the sewing operation, the present invention extracts the motion path of the robot arm according to visual perception, and proposes a robot arm sewing machine collaborative sewing method and system based on speed forward control. This method performs path planning before sewing and matches the sewing speed of each section, thereby avoiding the generation of wrinkles in advance, without strict requirements on fabric characteristics, etc.

[0071] When matching: Calculate the curvature, divide the path into different segments, set the approximate speed of the straight segment according to the human operation speed, and adjust the speed of the parts with different curvatures on this basis. Determine the speed of each segment of the robot arm. During movement, predict and adjust the error between this speed and the actual operation to obtain the speed of the robot arm. Then send the corresponding control signal to the sewing machine to make the speeds of the two consistent, and then match.

[0072] Embodiment 1

[0073] See attached Figure 1 As shown, this embodiment discloses a robot collaborative sewing system based on speed forward control, comprising:

[0074] The system consists of five modules: image detection module, pressure detection module, sewing equipment, controller and robotic arm. The image detection module of the system uses the Realsence depth camera to collect fabric image information in real time, including color images and depth images of the sewing environment, and directly obtains the shape and position characteristics of the fabric;

[0075] The pressure detection module uses a six-dimensional force sensor to obtain the pressure of the end pressing the fabric in real time during the sewing process, and maintains the end pressure to prevent the fabric from sliding;

[0076] After processing the image information, the controller obtains the shape and position characteristics of the fabric, further processes the path information and speed information, and outputs it to the robot arm and sewing equipment to control the two to sew the fabric in coordination. The controller outputs a speed to the robot arm and a speed to the sewing machine to keep the size and direction of the two consistent.

[0077] Attached Figure 1 In the figure, the data feedback of pressure detection is the pressing height, which is sent to the controller to determine the end position of the robot arm (the figure directly points to the robot arm because it is ultimately acting on the robot arm). The end position of the robot arm is a series of coordinates. These coordinates (x, y, z, rx, ry, rz) include position and rotation, which are given by the controller. After planning the speed of the robot arm, send the corresponding voltage signal to the sewing machine to control the speed for speed matching. Among them, the relationship between voltage and sewing machine speed can be calculated according to the sewing machine parameters.

[0078] Embodiment 2

[0079] See attached Figure 2 As shown, this embodiment discloses a robot collaborative sewing method based on speed forward control. Collaborative sewing mainly refers to speed coordination, that is, speed matching. The robot arm speed is planned and the sewing machine speed is set accordingly, including: contour extraction, trajectory planning, operation execution, and pressure control.

[0080] First, the image of the fabric to be sewn is collected, and the motion path of the end of the robot arm is extracted from the fabric contour through hand-eye calibration and matrix calculation. Then, according to the motion path and speed requirements, speed forward processing is performed and the trajectory is precisely interpolated to ensure high accuracy and stability, thereby achieving collaborative sewing. The specific steps are as follows:

[0081] Step 1: Use the Realsence depth camera to collect the color image and depth image of the global sewing environment. In this step, the camera field of view should include the entire fabric to obtain the information of the overall sewing environment;

[0082] Step 2: Process the image, extract the image contour and perform rough interpolation to obtain a contour point sequence.

[0083] Step 3: Based on the hand-eye calibration, the pixel coordinates are converted to the robot arm coordinate system through matrix operations, and the pressure point position and motion path of the end are generated according to the contour points.

[0084] Step 4: Introduce speed look-ahead control, optimize the speed extreme value according to the motion path, and complete the speed planning of the entire trajectory. Perform fine interpolation on the planned trajectory again to improve trajectory accuracy and ensure stable speed.

[0085] Step 5: Control the movement of the robot arm according to the fine interpolation result, and send start / stop and speed instructions to the sewing machine. Use the force sensor to detect the force data in real time and transmit it to the controller for PID control to maintain the Z-axis pressing force F. z , achieving collaborative sewing with the robotic arm.

[0086] In this step, the height of the end of the robot arm is controlled by PID. If the pressure exceeds the set data, the end of the robot arm is raised to prevent the fabric from being dragged due to excessive friction. If the pressure is less than the set data, the end of the robot arm is lowered to prevent the end from not contacting the fabric and being unable to press.

[0087] In this implementation example, in step 2, the process of extracting the image contour and performing rough interpolation to obtain a contour point sequence is as follows:

[0088] Among them, one of the color image and the depth image contains color information, and the other contains depth information. When processing, the information of the two images is first matched, that is, each pixel corresponds to a depth, which is convenient for subsequent acquisition and use of this information.

[0089] Step 2-1) Generate a point cloud file using the depth image and save it; read the color image, depth image and point cloud file.

[0090] Step 2-2) Define the HSV range according to the color of the fabric, filter the image, and the filtered image only contains the approximate image of the fabric. Erode and dilate the filtered image, and use the Canny operator to detect the edge of the fabric to obtain the outline of the fabric.

[0091] HSV (hue, saturation, value) is a color model used to filter the required colors.

[0092] Step 2-3) Perform equidistant sampling on the detected contour lines to form a dense point set (q0, q1, q2, ...q n ). The dense point set is a series of points obtained by initially sampling the contour, q i represents the sampling point;

[0093] Step 2-4) Fit the NURBS curve based on the dense point set obtained by sampling, which is defined as:

[0094]

[0095] Among them, N i，q (u) is the B-spline basis function, q i is the control point, w i is the weight, u is the curve parameter. C(u) is the point on the curve, is the coordinate at parameter u, q is the order of the curve (different from qi)

[0096] Step 2-5) Select a suitable step length Δu, starting from u=0, and calculate the corresponding C(u) point every other step length Δu to obtain a new interpolation point sequence (p0, p1, p2, ... p n ).

[0097] Step 2-6) generates a point cloud image from the interpolated point cloud to form a 3D outline of the cloth.

[0098] In step 3, the process of generating the end point position and motion path is as follows:

[0099] Step 3-1) Convert the interpolated coordinate points from the camera coordinate system to the robot coordinate system as a point set (P0, P1, P2, ...P n ).

[0100] Step 3-2) Identify the x coordinates of the point max 、x min and max ,y min , take the average value as the pressing point

[0101] Coordinate identification: traverse this point set (P0, P1, P2, ...P n ) coordinates to select the maximum and minimum values.

[0102] 3.3 For adjacent cloth contour points P i and P i+1 , calculate the direction angle Get the rotation angle and then calculate the end P end To contour point P i The translation vector of . Combining the two, a transformation matrix is ​​constructed to transform the contour point coordinates into the end path.

[0103] Transformation matrix: Transforming from one coordinate system to another requires displacement information and rotation information. In 3-2, after the pressing point is determined, each contour point has a fixed position relationship to the end pressing point. Based on this translation and rotation relationship, a matrix can be obtained.

[0104] The sewing machine needle is fixed, and when each contour point moves under the needle, the coordinates of the end can be calculated based on this fixed relationship matrix.

[0105] In this implementation example, in step 4, the speed look-ahead control process is as follows:

[0106] 4-1) Constrain speed and acceleration: Set the minimum speed v according to the manual sewing speed range min and minimum acceleration a min , to ensure smooth and continuous operation; set the maximum speed v max and the maximum acceleration amax To prevent equipment vibration and shock.

[0107] 4-2) Based on the point set obtained in step 3, calculate the curvature value k of each path point i Using the discrete curvature formula, for each point P on the path i , through its front and rear points P i-1 and P i+1 The coordinates are calculated:

[0108]

[0109] 4-3) Define a sliding window, calculate the curvature values ​​of n consecutive points in the path, and take the average rate of change to detect mutations:

[0110]

[0111] Set the threshold k threshold ,like It is believed that P i+n-1 Speed ​​mutation point.

[0112] 4-4) Obtain the velocity mutation point P according to the above method j and P k , calculate the path length between mutation points:

[0113]

[0114] 4-5) Confirm the number of forward-looking segments and first calculate the average distance between two adjacent interpolation points:

[0115]

[0116] Then calculate the number of forward-looking segments: Make sure to plan for enough points.

[0117] The number of look-ahead segments refers to the number of path points that need to be predicted and adjusted in advance.

[0118] 4-6) When the speed changes greatly before and after the mutation point and the path is short, the acceleration requirements may not be met, so the speed needs to be corrected.

[0119] Places with small changes in curvature (such as straight lines) can maintain a constant speed for a relatively long path. In contrast, a short path refers to places with large changes in curvature and where the speed needs to be adjusted multiple times.

[0120] The more speed mutation points there are, the shorter the path is (the speed needs to be adjusted multiple times); the fewer speed mutation points there are, the longer the path is.

[0121] If in the acceleration stage, from v i Accelerate to vi+1 , need to correct v i+1 , so that v′ i+1 ≤v i+1 .

[0122] If it is in the deceleration stage, from v i Decelerate to v i+1 , need to be repaired i , so that v' i ≤v i .

[0123] When processing short planning elements, before planning long segment i, determine whether the i+1 segment is a short segment. If so, continue to index backward until the i+k+1 segment is found to be a long segment. At this time, reversely adjust the feed speed from the i+k segment to the i+1 segment, and then forward plan the terminal speed of the i-th curve segment, and complete the planning of each larger part of the curvature in turn.

[0124] Short planning elements are places where the curvature changes greatly and the speed needs to be adjusted multiple times. The speed changes between shorter paths and needs to be adjusted.

[0125] 4-7) Based on the speed and acceleration limits, as well as the skilled manual operation speed, NURBS curves are used to interpolate the feed speed and position information in real time, inserting more points in the curve segments to ensure a smooth transition; and reducing the number of points in the straight line segments to keep the path simple.

[0126] Use NURBS curves to interpolate feed speed and position information in real time: Use the NURBS interpolation formula used in 2-4 above, set different parameters u for the straight line and curve parts, and obtain interpolation points of different densities.

[0127] According to the speed and acceleration constraints and the manual operation speed, the corresponding speed is preset. In actual operation, according to the error between the actual speed and the preset speed, the speed of the subsequent path is dynamically calculated to correct the speed and adjust the sewing machine speed at the same time.

[0128] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0129] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A robot collaborative sewing system based on speed forward control, characterized in that: include: Image detection module, pressure detection module, sewing equipment, controller and robotic arm; The image detection module collects image information of the fabric in real time and transmits it to the controller; The pressure detection module obtains the pressing height of the end pressing the fabric in real time during the sewing process and feeds it back to the controller; After processing the image information, the controller generates the pressing point position and motion path of the end, introduces speed look-ahead control, optimizes the speed extreme value according to the motion path, completes the speed planning of the entire trajectory, interpolates the planned trajectory, and obtains the interpolated result; The controller then determines the position and posture of the end of the robot arm according to the interpolated result and the pressing height; The robot arm performs actions to the fabric based on the end position; The controller sends a control signal to the sewing device, and the sewing device performs cloth feeding and sewing; The sewing device matches the speed of the robot arm to perform coordinated sewing.

2. A robot collaborative sewing system based on speed look-ahead control as claimed in claim 1, characterized in that: The image detection module is a depth camera, and the image information of the fabric collected in real time is a color image and a depth image of the sewing environment.

3. A robot collaborative sewing method based on speed forward control, characterized in that: include: Collecting image information of fabric; Obtain the pressing height of the fabric at the end during sewing in real time; After processing the image information, the pressing point position and motion path of the end are generated, and the speed look-ahead control is introduced. According to the motion path, the speed extreme value is optimized to complete the speed planning of the entire trajectory, and the planned trajectory is interpolated to obtain the interpolated result; Determine the end position of the robot arm according to the interpolated result and the pressing height; The robot arm performs actions to the fabric based on the end pose; Sewing equipment is used for feeding and sewing; The sewing equipment matches the speed of the mechanical arm to perform coordinated sewing.

4. A robot collaborative sewing method based on speed look-ahead control as claimed in claim 3, characterized in that: The image information is processed, including: Extract image contours from image information and perform rough interpolation to obtain a sequence of contour points; Based on the calibration, the pixel coordinates are converted to the robot coordinate system through matrix operation, and the pressing point position and motion path of the end of the robot are generated according to the contour points.

5. The robot collaborative sewing method based on speed look-ahead control as claimed in claim 3 is characterized in that: The image contour is extracted from the image information and rough interpolation is performed to obtain the contour point sequence. The specific process is as follows: The image information includes a color image and a depth image of the sewing environment; Generate 3D point cloud images using depth images and save them; read RGB images, depth images and point cloud images; Define the HSV range according to the color of the cloth, filter the image, erode and dilate the image, and use the Canny operator to detect the edge of the cloth; Sampling the detected contours to form a dense point set; Fitting NURBS curve according to sampling points; Select a suitable step length Δu, starting from u=0, and calculate the corresponding C(u) point every other step length Δu to obtain a new interpolation point sequence; The interpolated point set is used to generate a point cloud image to form the 3D outline of the cloth.

6. The robot collaborative sewing method based on speed look-ahead control as claimed in claim 3 is characterized in that: The specific process of generating the pressing point coordinates and the end path is as follows: The interpolated coordinate points are converted from the camera coordinate system to the robot arm coordinate system as a point set through matrix operations; Identify the x coordinates of the point max 、x min and max ,y min , take the average value as the pressing point; For adjacent cloth contour points P i and P i+1 , calculate the direction angle, obtain the rotation angle, and then calculate the end P end To contour point P i The translation vector of Construct a transformation matrix to transform the contour point coordinates into the end path.

7. A robot collaborative sewing method based on speed forward control as claimed in claim 3, characterized in that the speed The specific process of forward-looking control is as follows: Constrain velocity and acceleration: set a minimum velocity v min and minimum acceleration a min , set the maximum speed v max and the maximum acceleration a max ; According to the obtained point set, calculate the curvature value k of each path point i Using the discrete curvature formula, for each point P on the path i , through its front and rear points P i-1 and P i+1 Calculate the coordinates of Define a sliding window, calculate the curvature values ​​of n consecutive points in the path, and take the average rate of change to detect mutations; Set the threshold k threshold ,like It is believed that P i+n-1 Speed ​​mutation point. According to the above method, the adjacent velocity mutation point P is obtained. j and P k , calculate the path length between mutation points: To determine the number of forward-looking segments, first calculate the average distance between two adjacent interpolation points: Then calculate the number of forward-looking segments based on the average value to ensure that enough points are planned; When the speed changes greatly before and after the mutation point and the path is short, the acceleration requirement may not be met and the speed should be corrected; Use NURBS curves to interpolate feed speed and position information in real time.

8. The robot collaborative sewing method based on speed look-ahead control as claimed in claim 3 is characterized in that: When using NURBS curves to interpolate feed speed and position information in real time, insert more points in the curve segment to ensure a smooth transition; reduce the number of points in the straight line segment to keep the path simple.

9. The robot collaborative sewing method based on speed look-ahead control as claimed in claim 3, characterized in that: Correct the speed, specifically: If in the acceleration stage, from v i Accelerate to v i+1 , need to correct v i+1 , so that v' i+1 ≤v i+1 ; If it is in the deceleration stage, from v i Decelerate to v i+1 , need to be repaired i , so that v′ i ≤v i .

10. The robot collaborative sewing method based on speed look-ahead control as claimed in claim 3, characterized in that: When processing short planning elements, before planning the long segment i, determine whether the i+1th segment is a short segment. If so, continue to index backward until the i+k+1th segment is found to be a long segment. At this time, adjust the feed speed in reverse from the i+kth segment to the i+1th segment, and then forward plan the terminal speed of the i-th curve segment, and complete the planning of the larger parts of each curvature in turn.

Images