Systems and methods for sewing and smoothing fabric

Abstract

The automated robotic sewing system includes a sewing device, a work surface on which fabric is sewn by a sewing machine, and a dual-arm manipulator robot equipped with a force / torque sensor and a fabric handling end effector for moving the fabric across the work surface. A vision module is positioned above the work surface and above the robot manipulator. The vision module provides an image of the fabric within at least the area where the fabric is to be sewn, and a controller sequences the sewing device, robot manipulator, and vision module to enable seam line determination, seam line tracking, and wrinkle removal under visual servo control. As a result, freeform sewing and smoothing of the fabric is achieved.

Description

[Technical Field]

[0001] The present invention relates to the field of garment production, and more particularly to an automated sewing system having a dual-arm manipulator robot with end effectors for manipulating pieces of fabric to facilitate the sewing process. [Background technology]

[0002] In the field of garment production, pieces of fabric or fabric panels are manually handled and transferred to a sewing workstation for the intended seaming, hemming, or backstitching process. In the traditional garment production industry, the sewing process is primarily carried out in a manner where the orientation or shape of the fabric pieces is manually controlled with the aid of prefabricated fixtures or fasteners so that the fabric can be sewn along the desired seam line. Such a process is labor-intensive, and the accuracy, uniformity, and speed of the stitching are largely dependent on the ability of the operator himself.

[0003] Thus, in a traditional sewing process, workers use their hands to precisely manipulate pieces of fabric and sew them along a desired seam line using a sewing machine in a typical garment production process, such as piecing or backstitching. Precision sewing requires skilled techniques acquired after considerable periods of training and practice, and the manual sewing process itself is labor-intensive, i.e., not efficient production, and prone to difficulties brought about by global labor shortages. While some sewing tasks can be automated by machines, such automated sewing machines require specialized fasteners and fixtures that must be custom designed and fabricated for each shape, size, and material of the fabric piece. The primary purpose of these fasteners and fixtures is to hold the fabric pieces in place and provide a suitable frame or groove along which the sewing task is guided. While such approaches have the potential to automate certain sewing tasks that require less human intervention, their adaptability is hindered by the fact that the shapes of the fasteners and fixtures are all fixed and cannot be adapted to shapes other than those for which they are designed. While this automation approach is applicable to mass production of clothing products with the same structure and style, changing the style requires replacing all of the fastenings / clamps and reconfiguring the machine, making such an approach impractical for the production of highly customized clothing, which the market, especially the online retail sector, sees great potential in. Therefore, such equipment is labor-intensive and not suited to the production of customized clothing, thus calling for a new type of production methodology that allows for "high-mix, low-volume" manufacturing at a reasonable cost.

[0004] Considering the case of manual sewing, the operator not only needs to manipulate the fabric pieces relative to the sewing machine to ensure that the desired stitching path or stitch line is precisely followed, but also needs to ensure that the fabric pieces are properly stretched and, among other things, that there are no undesired folds or wrinkles in the areas to be sewn together. Failure to do so will result in an interruption of the sewing process and / or a reject product. Therefore, the ability to detect folds and wrinkles and, if necessary, eliminate them from the fabric represents a key feature of an automated sewing system that can perform the sewing process in the same intelligent and flexible manner as a human operator.

[0005] While some semi-automated systems have become available on the market to automate the sewing of certain classes of fabric elements, these systems still rely on the use of fixtures / fasteners, thereby limiting production adaptability for wider customization of clothing products. Some of these conventional systems use computer vision and robotic arms to manipulate the fabric pieces during the sewing process. See, for example, U.S. Pat. No. 10,366,175. However, many require the use of frames or fixtures, which limits the types of garments that can be produced with a single device. They may also require 3D body scanning, which utilizes ink fiducial marks on the fabric as part of the computer vision. Some prior art systems are also capable of detecting and smoothing out wrinkles. See, for example, U.S. Pat. No. 11,053,618, which uses motion devices to (i) transport the fabric pieces in one direction, (ii) rotate the fabric around an axis, or (iii) twist / stretch / flex the fabric to reduce wrinkles. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] U.S. Patent No. 10,366,175 [Patent Document 2] U.S. Patent No. 11,053,618 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention is directed to an automated robotic sewing process that uses an intelligent system capable of precise manipulation of fabric materials to solve the problems of the prior art, namely: (1) a wide range of sewing tasks must be performed repetitively by hand, while factories are facing a global shortage of garment workers; and (2) mechanical fixtures / fasteners, which are essential in today's sewing systems, must be designed and manufactured based on each and every different shape and material of the fabric piece. [Means for solving the problem]

[0008] The conceptual model of the automated robotic sewing system of the present invention has several main features, including (i) a sewing device, (ii) a dual-arm manipulator robot equipped with force / torque sensors and fabric handling end effectors, (iii) a vision module, and (iv) a sequence that enables seam line determination, seam line tracking, and wrinkle removal. This sewing robot system is a universal sewing automation system that does not require any auxiliary fasteners or fixtures that are otherwise required in conventional automation equipment to assist the sewing process. The present invention can be applied to sewing a single piece of fabric, or multiple pieces of fabric, along pre-set seam lines of various shapes, ranging from nearly straight lines to complex lines with various curvatures.

[0009] The seam line to be sewn may be a visible line or an invisible line. If it is a visible line, it may be a line marked / printed directly on the fabric material or projected as an image onto the fabric material. If it is an invisible line, it may be a virtual line offset by a distance (i.e., margin) from the physical edge or boundary of the fabric material in question, or any other free-form line as determined by its corresponding computer-aided design (CAD) model.

[0010] In the robotic manipulator-based system of the present invention, once a piece of fabric is brought under the robotic system's field of view, the system can automatically manipulate the piece of fabric and sew along the desired seam line using a sewing device. The seam line can be either a visible line marked on the surface of the piece of fabric or a virtual line determined by computer-generated data such as a CAD model. By utilizing a dual-arm manipulator configuration, the robot is capable of coordinated operation of two robotic arms, functioning similarly to the upper limbs of a human worker during the fabric manipulation process. The system's computer vision can accurately track the seam line and analyze images of the target piece of fabric for surface irregularities. When wrinkles or creases are identified, the robot takes corrective action to flatten the fabric. Automated sewing using the present invention can effectively replace manual processes and enable the ad-hoc, customized production of garments without the use of any fixtures or fixtures.

[0011] With the present invention, computer vision guidance and wrinkle detection can be achieved with only 2D video from a monochrome camera providing grayscale images.

[0012] The patent and application file contains at least one drawing executed in color. Copies of this patent and patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0013] The above and other objects and advantages of the present invention will become more apparent when considered in conjunction with the following detailed description and the accompanying drawings in which like designations indicate like elements in the various figures. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a diagram of an exemplary mechanism for a robotic sewing and smoothing system according to the present invention. FIG. [Figure 2] 1 is a structural diagram of a robotic sewing and smoothing system according to the present invention; [Figure 3] FIG. 1 is a diagram of a base sequence for an exemplary robotic sewing and smoothing system according to the present invention. [Figure 4] 10A and 10B are diagrams illustrating an example of a stitch line determination sequence according to the present invention. [Figure 5] FIG. 1 is a diagram of an exemplary mechanism for automated sewing under visual seam line tracking control in accordance with the present invention. [Figure 6] FIG. 10 is a diagram of an example of a visual seam line tracking control sequence according to the present invention. [Figure 7] 7A and 7B are illustrations of a piece of fabric sewn by the developed system of the present invention, in which Fig. 7A is a diagram of a straight line, Fig. 7B is a diagram of an arc line, and Fig. 7C is a diagram of a wavy line. [Figure 8] 1 is a diagram of an exemplary mechanism for automated smoothing under visual anti-wrinkle control according to the present invention. [Figure 9] FIG. 10 is a diagram of an example of a visual wrinkle prevention control sequence according to the present invention. [Figure 10] 10A and 10B illustrate the removal of wrinkles on fabric by a robotic system using visual wrinkle prevention control according to the present invention, where FIG. 10A shows the initial state of the wrinkled fabric and FIG. 10B shows the final state where the wrinkles have been removed. DETAILED DESCRIPTION OF THE INVENTION

[0015] Figure 1 illustrates one embodiment of the present invention for automated freeform sewing and smoothing under visual servo control. The system consists of two robotic manipulators 11 (i.e., dual-arm manipulators), each equipped with a force / torque sensor 13 and an end effector 15 for moving a textile material 10 across a work surface 12. The system further comprises a camera (vision sensor) 14, a projector (pattern generator) 16, and a sewing machine 18. Operation of the system is controlled by a controller 17, e.g., by a computer as described further below.

[0016] The end effectors 15 are used to move the fabric material 10 and may involve different designs and mechanisms depending on the complexity of the task. For example, they may be multi-fingered grippers, specialized end effectors for freeform folding and / or pick-and-place processes (such as the F-FOLD end effector disclosed in PCT / CN2022 / 090459), vacuum heads, or pads made of soft / elastic material. In this embodiment, a simple flexible foam pad or sponge without an internal actuator is used as the end effector for manipulating the fabric.

[0017] As the process begins, the dough is pushed down by two end effectors whose movement is controlled by the robotic manipulator. Force / torque sensors between the robotic manipulator and the end effectors sense the push-down force and help adjust the force and position of the end effectors relative to the work surface. This prevents the end effectors from slipping as the dough is held down and translated across the surface. Note that while the system can operate with a single robotic manipulator, a dual-arm manipulator robot, such as that depicted in Figure 1, is a preferred configuration, as the coordinated movement of the robotic arms provides increased efficiency in performing the same task, such as smoothing or stretching dough.

[0018] A sewing device or sewing machine 18 is an integral part of the system. The sewing machine utilized in this embodiment is a programmable sewing machine whose operation can be synchronized with the robot and generally controlled by the system's central controller. Nevertheless, the use of a programmable sewing machine is not required, and most off-the-shelf sewing machines can be configured with the robot system. Modern sewing machines generally feature a sewing needle, a presser foot, and a feed dog. The function of the presser foot is to apply pressure to the top of the fabric to press it against the feed dog, while the purpose of the feed dog is to grip the bottom portion of the fabric and help advance it through the sewing machine.

[0019] Because fabric movement is fully controlled by the robot in this system, the presser foot and feed dog of the sewing machine are not necessary and can simply be removed or disengaged. For example, most modern sewing machines allow the feed function to be stopped by raising the presser foot and lowering the feed dog. This allows these sewing machines to be integrated with the robotic system for automated sewing on a work surface approximately flush with the sewing machine's needle plate. In a more preferred configuration, the sewing device in this invention is a programmable sewing machine in which the system's central controller 17 (e.g., a computer equipped with an x86-based multicore processor, memory, image processing, I / O ports, and network interfaces) can independently control the sewing needle speed. A key advantage of a programmable sewing machine is that it can change the needle speed in response to any changes in the translational speed of the end effector that feeds the fabric, thereby ensuring a uniform stitch pitch along the stitch line. Without real-time control of the needle speed, as is the case with non-programmable sewing machines, the fabric feed speed must be maintained constant by the robot.

[0020] Located above the end effector 15 is a vision module comprising a camera unit 14 and a projector unit 16 with an illumination source. The camera serves to perform image acquisition, while the projector provides an illumination pattern on a surface area that completely or partially covers the target fabric. An exemplary configuration of the robotic system is provided in FIG. 2. The system may include one or more of each type of system component as shown, including a robotic manipulator 11, a force / torque sensor 13, an end effector 15, a sewing device (e.g., a sewing machine) 18, a camera 14, a projector 16, and an auxiliary actuator 20. The auxiliary actuator 20 can be used to assist in handling the fabric piece or to toggle a switch / button to activate additional machinery. A central controller 17 connects the system components and implements several algorithms 22, including "seam line determination," "visual seam line tracking control," and "visual wrinkle prevention control." It should be noted that the term "stitch line" in this context has the same meaning as "stitch path" or "sewing path" and is not limited to processes in which multiple pieces of fabric are sewn together. It can also be applied to embroidery processes in which only a single piece of fabric is processed.

[0021] The basic sequence of the robotic sewing and smoothing workflow is illustrated in Figure 3. After initializing the system components (step 301), the central controller 17 analyzes images of the target fabric 10 taken or acquired by the camera 14 (step 302) to verify whether a visible seam line is present on the fabric (step 303). If no seam line is present, the system executes a seam line determination sequence (step 304) to derive a seam line path by referencing a corresponding CAD model (or numerical data) for the current orientation and position of the fabric. Once a seam line is identified, the system performs automated sewing along that seam line under visual seam line tracking control (step 306) and visual wrinkle prevention control (step 305) until the job is completed (end). The key features of our method are: (1) independent control of internal and external forces acting on the fabric using force sensors; (2) no depth image (3D) is required; (3) only grayscale images are required; and (4) a feedback control scheme is used for both wrinkle removal and seam line tracking.

[0022] In the absence of a visible stitch line, e.g., a manually marked, printed, or projected line, a stitch line determination sequence allows for the calculation of a desired stitch line on a target fabric material. Figure 4 illustrates one example of such a sequence, beginning with initialization (step 401) and acquisition of a fabric image (step 402). A set of data points, such as a point cloud, is formed in space from the image (step 403) and compared to those of a corresponding CAD model of the target fabric piece (step 404). In doing so, the actual position and orientation of the fabric piece (fabric pose) is determined (step 405), allowing a virtual stitch line, i.e., sewing path, for the fabric piece to be calculated (step 406). A coordinate transformation is performed when a change in fabric pose is detected and the stitch line remains in the same location relative to the fabric piece. Once the seam is sewn, the program or algorithm returns to the base sequence (step 407). In contrast to the prior art U.S. Pat. No. 11,053,618, the present sequence does not require the collection of depth information or color (RBG) images, resulting in more relaxed requirements for cameras used with the present invention.

[0023] A mechanism for an automated sewing process under visual seam line tracking control is illustrated in Figure 5. Visual seam line tracking control works by visualizing a piece of fabric and controlling the operation of the actuators of the robotic system and sewing device so that the fabric is stably sewn along a preset seam line. Sensing devices included in this control method include a camera 14 and force / torque sensors connected to each end effector 15. The function of the force / torque sensors 13 is to monitor the forces exerted by the end effectors (i) vertically on the work surface and (ii) along a plane parallel to the work surface, i.e., laterally, on the fabric material. These forces are regulated by a closed-loop servo control, which adjusts the torque of each joint of the robotic manipulator 11 to achieve the desired force output.

[0024] During operation, images captured by camera 14 are analyzed in real time to calculate the trajectory of each end effector and the needle of the sewing device. The zone imaged by camera 24 should cover needle drop point 25 and at least a portion of the fabric material. Here, camera 14 may be a digital camera with a monochrome (e.g., grayscale) image sensor or a color image sensor, regardless of depth-sensing capability. The preset stitch line may be a visible line on the fabric, or it may be an invisible line determined by a stitch line determination sequence. By referencing the position data corresponding to the preset stitch line, the system calculates the pass point where needle drop point 25 coincides with the stitch line. In doing so, the angular velocities of the manipulator joints are determined, and the movement of the robot manipulator and the movement of the sewing needle are simultaneously controlled until the sewing operation of sewing along stitch line 26 is completed.

[0025] In another embodiment, the system can operate with conventional sewing machines whose needle speeds cannot be programmed by the system's controller. In this case, motion control is applied only to the joints of the robotic manipulator arm. A feedback control scheme can adapt to physical disturbances and errors associated with the calculated joint velocities, continuously tracking the stitch line relative to the needle drop point. An example sequence for visual stitch line tracking control is illustrated in Figure 6.

[0026] In Figure 6, the sequence begins with initialization (step 601) and acquisition (capture) of a fabric image (step 602). A calculation of the pass point where the needle drop point 25 coincides with the stitch line is performed in step 603. The angular velocities of the manipulator joints are then calculated in step 604. Based on the calculation of the pass point and the angular velocities of the joints, the motion control is determined (step 605). A determination is made in step 606 as to whether the sewing operation is complete. If not, the sequence returns to the beginning of step 603. If the operation is complete, the program or algorithm returns to the base sequence (step 607).

[0027] Figure 7 shows a series of sample fabric pieces (approximately 20 cm x 30 cm in size) being sewn along preset seam lines under visual seam line tracking control. In this illustration, the preset seam lines are printed on the fabric pieces for illustrative purposes. The results show that the sewing system is capable of automatically sewing fabric pieces along different preset shapes, including straight lines (left) (Figure 7A), arc lines (center) (Figure 7B), and wavy lines (right) (Figure 7C), with high positional accuracy.

[0028] An exemplary mechanism for visual wrinkle prevention control that enables automatic removal of wrinkles on fabric is illustrated in FIG. 8. For a smooth sewing process, appropriate measures should be taken to ensure that fabric pieces are free of folds and wrinkles before they are sewn. Therefore, in traditional manual sewing processes, human operators often stretch or tug on the fabric to flatten it before passing it through a sewing machine. The visual wrinkle prevention control of the present invention is an algorithm that enables the end effectors 15 to apply tension to the fabric along the appropriate direction to remove folds and wrinkles. The sensing devices included in this control module include cameras 14 and force / torque sensors 13 connected to each end effector 15. Here, the cameras can be digital cameras with RGB (color) or monochrome (e.g., grayscale) image sensors, regardless of depth-sensing capabilities. The zone imaged by the camera (imaging zone 24) should cover at least a portion of the fabric 10 where wrinkles are detected. The projector 16 can be any device capable of projecting a lighting pattern onto a surface. The force / torque sensors 13 serve to monitor the forces exerted by the end effector on the work surface in the normal direction and on the fabric material along a plane parallel to the work surface. Among other things, the force data is used as input to the visual servo control algorithm, which plays an important role in improving the stability, position accuracy and convergence range of the control system.

[0029] FIG. 9 shows a sequence illustrating the visual wrinkle prevention control method. It begins with initializing module execution in step 901. Because ridges and valleys in a fabric are difficult to recognize in a single image without triangulation, a pattern, preferably a structured pattern, is additionally projected onto the surface of the target fabric in step 902 to create additional contrast and is captured by the camera (step 903). The image data captured by the camera 14 is processed in real time and analyzed for deviations from a flat surface. To improve the field of view, the camera is preferably positioned in such a way that its optical axis is tilted at a fixed angle (preferably 45±20 degrees) from the surface normal of the target fabric. Wrinkles are identified by analyzing visual features from the image, without using three-dimensional topological data of the fabric, which requires distance or depth measurements (as in prior art U.S. Pat. No. 11,053,618). Data from force / torque sensors is also captured in step 903; these measure the various forces experienced by the end effector.

[0030] When the sequence determines that one or many wrinkles exist on the fabric, the angular velocity of each robot manipulator's joints is calculated (step 904), and the robot is operated to manipulate the fabric according to visual servoing techniques to smooth out the wrinkles in step 905. The stretching process involves pulling or stretching actions. When performing these actions, forces are measured to avoid excessive tension on the fabric, which could cause unintended damage. Additionally, data collected from the force / torque sensors is related to the tension generated within the fabric's structure and used as a feedback signal. If the force measured along the stretching direction changes rapidly, for example, when the fabric is sufficiently stretched, the stretching action is stopped. The above process is repeated until the system detects no more wrinkles on the fabric in step 906. The sequence then checks whether sewing is complete in step 907, and if so, returns to the base sequence (step 908). If not, the sequence returns to the beginning of step 906.

[0031] Figure 10 shows the process by which a piece of fabric (approximately 25 cm x 25 cm) is flattened by a robot under visual anti-wrinkle control. In this example, a camera captured images of the fabric illuminated with a checkerboard pattern by a projector. The checkerboard pattern serves as an example of a structured pattern. Other patterns are possible, provided the computer vision module is trained on these patterns. A force sensor attached to the end effector provides information on whether the fabric has been sufficiently stretched. The results demonstrate that wrinkles present on the fabric piece (left) were smoothed out by the coordinated movement of the end effector (right).

[0032] While the invention has been described in relation to particular embodiments, it is to be understood that various modifications thereof will become apparent to those skilled in the art upon reading the specification. It is therefore to be understood that the invention disclosed herein is intended to cover such modifications as fall within the scope of the appended claims.

Claims

1. An automatic robotic sewing system, comprising: Sewing machines and a work surface on which fabric is sewn by said sewing machine; at least one robotic manipulator arm, each comprising at least one force / torque sensor and at least one fabric handling end effector for moving the fabric across the work surface; a vision module positioned above the work surface and above the robotic manipulator, the vision module providing an image of the fabric at least in an area where the fabric is to be sewn; a controller that sequences the sewing machine, the robotic manipulator, and the vision module to enable seam line determination, seam line tracking, and wrinkle removal under visual servo control, thereby achieving freeform sewing and smoothing of the fabric.

2. 10. The automated robotic sewing system of claim 1, wherein the end effector is at least one of a gripper with multiple fingers, a specialized end effector for free-form folding or pick-and-place, a vacuum head, or a pad made of soft / elastic material.

3. 10. The automated robotic sewing system of claim 1, wherein the end effector is a simple flexible foam pad or sponge and does not include an internal actuator.

4. 2. The automated robotic sewing system of claim 1, wherein the vision module includes a camera operable to perform image acquisition and a projector having an illumination source to provide an illumination pattern onto a surface area covering the fabric, both of which are part of the visual servo control.

5. 5. The automated robotic sewing system of claim 4, wherein the camera is a monochrome camera that produces grayscale images.

6. 10. The automated robotic sewing system of claim 1, further comprising an auxiliary actuator, which can be positioned to assist in handling the fabric or to toggle a switch / button when additional machine operation is required.

7. The automated robotic sewing system of claim 1 , wherein the processor connects the system components and implements a plurality of control algorithms.

8. The base algorithm is analyzing an image of the fabric taken by a camera; verifying whether there is a visible stitch line on the fabric, and if not, executing a stitch line determination sequence to determine a stitch line path with reference to a corresponding model for the current orientation and position of the fabric; and if the seam line is identified, performing automated sewing along the seam line under a visual seam line tracking control sequence and a visual anti-wrinkle control sequence until the job is completed.

9. The stitch lines are manually marked, printed, or projected from a projector of the vision module, and the stitch line determination sequence comprises: acquiring an image of the fabric; forming a set of data points in space from the image; determining the actual position and orientation of the fabric by comparing the data points with those of a corresponding model of the fabric; calculating a virtual seam line or a virtual sewing path for the fabric; performing a coordinate transformation when a change in the position and orientation of the fabric is detected and the stitch line remains in the same location relative to the fabric; and sewing the fabric.

10. The visual seam tracking control sequence includes: acquiring an image of the fabric; calculating a passing point where a needle drop point of the sewing machine coincides with the stitch line; calculating angular velocities of the joints of the robot manipulator; determining a motion control based on the calculation of the waypoints and the angular velocities of the joints; and determining whether the sewing operation is complete.

11. one of the algorithms is a visual wrinkle prevention control sequence; calculating angular velocities of joints of the robotic manipulator when the sequence determines that wrinkles are present on the fabric; controlling the robotic manipulator to manipulate the fabric according to visual servoing techniques to smooth out the wrinkles by pulling or stretching actions; measuring the force of the pulling or stretching action to avoid excessive stress on the fabric, which may cause damage; and checking to see if said wrinkles are still present and repeating said steps until said wrinkles are no longer present.

12. 12. The automated robotic sewing system of claim 11, wherein during the measuring step, data collected from the force / torque sensor is related to tension generated within the fabric structure and used as a feedback signal so that the stretching action is stopped when the force measured along the stretching direction changes rapidly, e.g., when the fabric is sufficiently stretched.

13. 10. The automated robotic sewing system of claim 1, further comprising a second robotic manipulator arm including a force / torque sensor and a fabric handling end effector for moving the fabric across the work surface.

14. 10. The automated robotic sewing system of claim 1, wherein the sewing machine preferably has at least one motor for actuating a sewing needle, which can be controlled by a processor to adjust stitch pitch.

15. 1. A method for automatically sewing fabric, comprising: analyzing an image of the fabric taken by a camera positioned above a work surface on which at least one piece of fabric is located; verifying whether a visible stitch line exists on the fabric, and if so, executing a stitch line determination sequence, and if not, deriving a path for the stitch line by referencing a corresponding model for the current orientation and position of the fabric; and once the seam line is identified, performing automated sewing along the seam line under a visual seam line tracking control sequence and a visual wrinkle prevention control sequence until the job is completed.

16. The stitch line determination sequence includes: forming a set of data points in space from the image; determining the actual position and orientation of the fabric by comparing the data points with those of a corresponding model of the fabric; calculating a virtual seam line or a virtual sewing path for the fabric; and performing a coordinate transformation when a change in the position and orientation of the fabric is detected and the stitch line remains in the same location relative to the fabric.

17. The visual seam tracking control sequence includes: calculating a passing point where a needle drop point of a sewing machine coincides with the stitch line; calculating angular velocities of the joints of the robot manipulator; determining a motion control based on the calculation of the waypoints and the angular velocities of the joints; and determining whether the sewing operation is complete.

18. The visual wrinkle prevention control sequence comprises: calculating angular velocities of joints of a robotic manipulator when the sequence determines that wrinkles are present on the fabric; controlling the robotic manipulator to manipulate the fabric according to visual servoing techniques to smooth out the wrinkles by pulling or stretching actions; measuring the force of the pulling or stretching action to avoid excessive stress on the fabric, which may cause damage; and checking to see if said wrinkles are still present and repeating said steps until said wrinkles are no longer present.

Images