Method and system for manufacturing apparel and textiles

Abstract

The system (11) includes: (i) an extrusion assembly (EA) (77, 78) configured to melt one or more granules of material (20) and eject the molten granules onto a substrate (44); (ii) a motion assembly (13) configured to move the EA (77, 78) relative to the substrate (44) along a first axis and oscillate the substrate along a second axis different from the first axis while the EA (77, 78) ejects the molten granules to generate a line of a filament (22) on the substrate (44); and (iii) a processor (66) configured to control the EA (77, 78) and the motion assembly (13) to position the filament (22) at a predetermined area of ​​the substrate (44).

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 63 / 317,984, filed March 9, 2022, the disclosure of which is incorporated herein by reference.

[0002] The present invention relates generally to fabric and garment manufacturing, and more particularly to methods and systems for manufacturing fabrics and garments using extrusion and machine learning techniques. [Background technology]

[0003] A variety of techniques have been published for producing fabrics and garments.

[0004] For example, US Patent Application Publication No. 2021 / 0025080 describes a bicomponent fiber including a first region formed from a condensation polymer and a second region formed from a polypropylene blend. The polypropylene blend includes (i) a propylene-based polymer having a density of 0.895 g / cm3 to 0.920 g / cm3 and a melt index I2 of 0.5 to 150 g / 10 min as determined by ASTM D1238 at 230° C. and 2.16 kg; (ii) a maleic anhydride grafted polypropylene; and (iii) an inorganic Brønsted-Lowry acid having an acid strength pKa value of 1 to 6.5 at 25° C., and the polypropylene blend has 0.03 to 0.3 weight percent of the grafted maleic anhydride based on the total weight of the polypropylene blend. The first region is a core region of the bicomponent fiber, and the second region is a sheath region of the bicomponent fiber, the sheath region surrounding the core region.

[0005] US 2014 / 0000784 describes a method for producing a multi-layer nonwoven web including a monocomponent fiber layer and a multicomponent fiber layer, the method including the steps of: spinning a plurality of monocomponent fibers on a beam of a monocomponent fiber spinneret and directing them to a belt to form a monocomponent batt layer; spinning a plurality of multicomponent fibers on a beam of a multicomponent fiber spinneret and directing them to a belt to form a multicomponent batt layer, the layers being disposed one on the other in close proximity to form a batt; and bonding the batts in a thermal bond pattern to form a multi-layer nonwoven web.

[0006] US Patent Application Publication No. 2021 / 0146606 describes printing apparatus and manufacturing techniques for the production and manufacture of pressure suits and space suits. The disclosure also relates to space suit components formed from additive polymers, and components including one or more layers of mesh fabric with an applied layer of additive polymer. One method includes providing an apparatus consisting of a six degree of freedom motion manipulator attached to one or more printing extruder heads. The additive polymer can be extruded through one or more printing extruder heads to form a 3D printed space suit component.

[0007] US Patent Application Publication No. 2019 / 0084196 describes articles for use in footwear and other applications, as well as related systems and methods. Certain embodiments relate to methods of making articles, such as depositing fibers onto a shoe last to form a nonwoven material. The fibers can be, for example, preformed fibers, or fibers including a reaction product of a composition extruded from one or more nozzles. In some embodiments, the method can include depositing fibers to form a nonwoven material using one or more nozzles, and 3D printing one or more features onto the nonwoven material. In some cases, articles including less than five pieces of material can be formed. In some embodiments, the article can include a nonwoven material including thermosetting fibers. Additionally, some embodiments relate to shoe lasts that can be suitable for forming nonwoven materials.

[0008] US Patent No. 10,696,034 describes a method, system, and apparatus for extrusion-based three-dimensional printing. The method, system, and apparatus enable printing materials such as fabrics, clothing, and wearable and / or implantable devices. Several different enhancements are provided that enable this improved form of three-dimensional printing, including (1) printing using a polymer (e.g., cellulose acetate) dissolved in a solvent (e.g., acetone); (2) selectively joining portions of a deposited filament onto one or more surfaces and / or one or more previously deposited filaments; (3) using a specific tool path to create a fabric or similar material by forming a woven pattern; and (4) printing across multiple layers even when the previous layer is not complete.

[0009] US Patent Application Publication No. 2011 / 0130063 (Patent Document 6) describes a spinning device capable of spinning fibers with a small fiber diameter stably and with good productivity, a nonwoven fabric manufacturing apparatus equipped with the same, a method for manufacturing a nonwoven fabric using the nonwoven fabric manufacturing apparatus, and a nonwoven fabric manufactured by the method. The spinning device has one or more liquid discharge outlets capable of discharging a spinning liquid, and one or more gas discharge outlets positioned upstream of each of the liquid discharge outlets and extending linearly, capable of discharging a gas, and can apply a shear force due to the gas and the associated airflow to the discharged spinning liquid in a single line. The nonwoven fabric manufacturing apparatus has a fiber recovery means in addition to the spinning device. The nonwoven fabric manufacturing method is a method using the nonwoven fabric manufacturing apparatus. The nonwoven fabric is a nonwoven fabric manufactured by this method. [Prior art documents] [Patent documents]

[0010] [Patent Document 1] U.S. Patent Application Publication No. 2021 / 0025080 [Patent Document 2] US Patent Application Publication No. 2014 / 0000784 [Patent Document 3] U.S. Patent Application Publication No. 2021 / 0146606 [Patent Document 4] US Patent Application Publication No. 2019 / 0084196 [Patent Document 5] U.S. Patent No. 10,696,034 [Patent Document 6] US Patent Application Publication No. 2011 / 0130063 Summary of the Invention

[0011] Embodiments of the invention described herein provide a system that includes: (i) an extrusion assembly (EA) configured to melt granules of one or more substances and project the molten granules onto a substrate; (ii) a motion assembly configured to move the EA relative to the substrate along a first axis and oscillate the substrate along a second axis distinct from the first axis while the EA projects the molten granules to generate a line of a filament on the substrate; and (iii) a processor configured to control the EA and the motion assembly to place a filament at a predetermined area of ​​the substrate.

[0012] In some embodiments, the EA is configured to generate one or more puddles of molten granules along the line. In other embodiments, the EA is configured to generate a section of the line by drawing a portion of the molten granules from the puddle and depositing the drawn portion on the substrate along a first axis. In still other embodiments, the processor is configured to (i) determine the predetermined region by controlling a first movement profile for moving the EA along the first axis and a second movement profile for oscillating the substrate along a second axis. In some embodiments, the processor is configured to control the motion assembly to move the EA in an opposite direction along the first axis to deposit additional filaments in additional predetermined regions of the substrate. In other embodiments, at least a portion of the predetermined region overlaps with at least a portion of the additional predetermined region. In still other embodiments, the processor is configured to control the motion assembly and the EA to generate a plurality of filaments and additional filaments and generate a first array of one or more bundles by bonding between two or more of the filament and additional filaments in each bundle.

[0013] In some embodiments, the processor is configured to control the motion assembly and the EA to generate at least one of the bundles by intertwining one or more of the filaments with one or more of the additional filaments. In other embodiments, the processor is configured to control the motion assembly and the EA to generate at least one bundle by bonding two or more of the filament and the additional filament to a puddle of molten granules. In yet other embodiments, the motion assembly is configured to rotate the substrate about an axis of rotation, and the processor is configured to (i) control the motion assembly to rotate the substrate through a predetermined angle of rotation, and (ii) control the motion assembly and the EA to generate one or more second arrays of bundles on the substrate.

[0014] In some embodiments, the processor is configured to control the motion assembly and the EA to generate a layer including at least two of the bundles of the first and second arrays intertwined with each other. In yet other embodiments, the processor is configured to control the motion assembly and the EA to generate the bundles intertwined in a crisscross configuration. In yet other embodiments, the processor is configured to control the motion assembly and the EA to generate the bundles intertwined in a first and second orientation, respectively, the first and second orientations defining a given angle between the intertwined bundles. In some embodiments, the processor is configured to control the motion assembly and the EA to generate one or more reinforcement points (RPs) between one or more pairs of the intertwined bundles. In other embodiments, the processor is configured to control the motion assembly and the EA to generate at least one of the RPs by welding at least one of the pairs together. In yet other embodiments, the processor is configured to control the motion assembly and the EA to generate at least one of the RPs by soldering between at least one bundle of the pairs.

[0015] In some embodiments, the processor is configured to control the motion assembly and the EA to (i) generate one or more layers and laminate the first plurality of layers to generate a first side of the fabric, and (ii) generate one or more additional layers and laminate the second plurality of additional layers to generate a second side of the fabric. In other embodiments, the first side has at least a first edge and the second side has at least a second edge, and the processor is configured to control the bonding assembly to bond between the first edge and the second edge to generate the fabric. In yet other embodiments, the processor is configured to control the robot to place a release layer between the first side and the second side to prevent undesired bonding between the first side and the second side. In some embodiments, the first side is disposed on the second side such that the first edge and the second edge are aligned, and the processor is configured to control the laser source to direct a laser beam to the aligned first edge and the second edge to bond between the first edge and the second edge. In other embodiments, the first surface is disposed over the second surface such that the first edge and the second edge are aligned, and the processor is configured to control a stitcher to stitch between the first edge and the second edge. In yet other embodiments, prior to being dissolved, the granules have a size between 1 μm and 10 mm.

[0016] In some embodiments, at least one of the granules is made from recycled textiles or recycled clothing. In other embodiments, at least one of the granules comprises a soluble polymer selected from the list of soluble polymers consisting of one or more of (i) polyester, (ii) nylon, and (iii) polypropylene. In yet other embodiments, at least one of the granules comprises a soluble polymer mixed with cotton. In some embodiments, the EA comprises a column and one or more screw extruders disposed within the column and configured to move the granules along the column. In other embodiments, the EA comprises one or more heating units configured to increase a temperature in a given section of the column to melt the granules. In yet other embodiments, the one or more heating units comprise (i) a first heating unit disposed in a first section of the column and configured to heat the granules to a first temperature to increase the flexibility of the granules, and (ii) a second heating unit disposed in a second section of the column different from the first section and configured to heat the granules to a second temperature higher than the first temperature to melt the granules.

[0017] In some embodiments, the EA includes one or more cooling ribs configured to dissipate heat generated in the first and second sections from heating the third section of the column. In other embodiments, the EA includes a motor configured to rotate at least one of the screw extruders to move the granules along the column. In yet other embodiments, the EA includes one or more nozzles, at least one of the nozzles configured to generate lines by applying molten granules onto a substrate.

[0018] In some embodiments, the nozzle includes (i) a first nozzle configured to generate a first line in a first region of the substrate, and (ii) a second nozzle configured to generate a second line in a second region of the substrate different from the first region. In other embodiments, the EA includes an air curtain configured to perform one or both of (i) shaping the line of filaments generated on the substrate, and (ii) improving the attachment of the line of filaments to the substrate. In still other embodiments, the EA includes a mechanism for increasing a flow rate of the molten granules to generate puddles along the line. In some embodiments, the mechanism includes a spring. In other embodiments, the EA includes a machine learning (ML) engine configured to receive input including one or more attributes of one or more physical properties of the textile and output a mechanical structure of the textile. In still other embodiments, the mechanical structure includes one or more of (i) one or more of the filaments, (ii) one or more puddles, (iii) one or more bundles generated by intertwining a plurality of filaments, and (iv) one or more layers including one or more arrays of intertwined first bundles and second bundles.

[0019] In some embodiments, the ML engine is configured to output a recipe for generating the mechanical structure in the system. In other embodiments, the ML engine is implemented in software and the processor is configured to execute the software. In yet other embodiments, the fabric includes a first fabric section having a first physical characteristic and a second fabric section having a second physical characteristic different from the first fabric section and different from the first physical characteristic, and the ML engine is configured to output a first mechanical structure of the first fabric section and a second mechanical structure of the second fabric section different from the first mechanical structure. In some embodiments, the paddles are formed at random positions within the predetermined area. In other embodiments, the paddles include (i) a first paddle for bonding together a first number of filaments and (ii) a second paddle for bonding together a second number of filaments different from the first number.

[0020] There is further provided, in accordance with an embodiment of the present invention, a method comprising melting one or more granules of a material in an extrusion assembly (EA) and projecting the molten granules onto a substrate, the EA moving the EA relative to the substrate along a first axis and the substrate oscillating along a second axis different from the first axis while the EA projects the molten granules to generate a line of filaments on the substrate.

[0021] In some embodiments, the method includes generating one or more puddles of molten granules along a line, while in other embodiments, the method includes generating a section of the line by drawing a portion of the molten granules from the puddle and disposing the drawn portion on a substrate along a first axis.

[0022] The present invention will be more fully understood from the following detailed description of the embodiments thereof, taken in conjunction with the drawings. [Brief description of the drawings]

[0023] [Figure 1] 1 is a schematic diagram of a system for manufacturing fabrics and / or garments according to an embodiment of the present invention.

[0024] [Figure 2A] 2 is a schematic front view of an extrusion assembly (EA) of the system of FIG. 1 in accordance with an embodiment of the present invention.

[0025] [Figure 2B] 2 is a schematic side view of an EA of the system of FIG. 1 in accordance with an embodiment of the present invention.

[0026] [Diagram 3] FIG. 2 is a schematic diagram of the arrangement of bundles in each fabric layer according to an embodiment of the present invention.

[0027] [Figure 4] FIG. 2 is a schematic diagram of the arrangement of bundles in each fabric layer according to an embodiment of the present invention.

[0028] [Diagram 5] FIG. 2 is a schematic diagram of the arrangement of bundles in each fabric layer according to an embodiment of the present invention.

[0029] [Figure 6] FIG. 2 is a schematic diagram of the arrangement of bundles in each fabric layer according to an embodiment of the present invention.

[0030] [Figure 7] FIG. 1 is a block diagram that illustrates generally a machine learning based system configured to automatically determine a mechanical structure of a fabric based on physical properties of the fabric, in accordance with an embodiment of the present invention.

[0031] [Figure 8] 2 is a schematic diagram of a garment produced by the system of FIG. 1 in accordance with an embodiment of the present invention.

[0032] [Figure 9] 1 is a flow chart that generally illustrates a method for generating garments, according to an embodiment of the present invention.

[0033] [Figure 10] 2 is an image of an EA of the system of FIG. 1 according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] overview

[0035] The embodiments of the present invention described below provide improved techniques for manufacturing fabrics and garments.

[0036] Typically, garments such as shirts (also referred to herein as clothing) are manufactured in three facilities: (i) textile mills where fibers are produced, (ii) fabric mills where fibers are converted into fabric, and (iii) garment factories where the fabric is cut and shaped and then the shaped fabric is sewn or stitched to produce the garment. The formation of shirts in such facilities has several disadvantages, including, but not limited to, high waste (gas, liquid, and solid) emissions and effluents, being labor intensive, and having long cycle times.

[0037] Moreover, such processes have limited flexibility, e.g., (i) they are suitable for low product mix and large unit quantities of each product, and (ii) each modification or customization increases the cycle time and cost of producing a garment. Moreover, the recycling rate of garments and fabrics produced by such processes is negligible, increasing the amount of waste accumulated on the planet.

[0038] Thus, there is a strong demand for fabric and garment production techniques that provide increased flexibility, customizability, and environmental friendliness, among other things.

[0039] In some embodiments, a system for producing fabrics and garments includes an extrusion assembly (EA), a motion assembly, and a processor. The EA is configured to receive granules of one or more substances produced from recycled garments or any other source. The EA is configured to move the granules along a column, melt the granules, and eject the molten granules onto a moving substrate.

[0040] In some embodiments, the motion assembly is configured to move the EA relative to the substrate along a first axis (e.g., along the Y axis of an XYZ coordinate system) and vibrate the substrate in a second axis different from the first axis. For example, the vibration can be performed along the X axis of an XYZ coordinate system, or in a direction that is a combination of the X and Y axes.

[0041] In such an embodiment, the combined motion of the EA and the substrate in two different axes generates a line of filaments by the EA that affixes molten granules to the substrate. The line may have an irregular zigzag shape to intertwine multiple filaments into a bundle. It is noted that the motion assembly is configured to vibrate the substrate using any suitable frequency or suitable combination of frequencies. For example, the processor is configured to control the motion assembly to vibrate the substrate with a random motion profile in which the frequency or combination of frequencies varies over time.

[0042] In some embodiments, the EA is configured to generate one or more puddles of molten granules along a line. The positions of the puddles along the line may be random along the line or may be at predetermined positions. Further, the EA is configured to generate a section of the line by drawing a portion of the molten granules from the puddle and depositing the drawn portion on the substrate along a first axis (e.g., along the Y axis).

[0043] In some embodiments, the processor is configured to control the EA and the motion assembly to place the filament and one or more paddles at a predetermined region of the substrate, After generating the line of filament, the processor is configured to control the motion assembly to reverse the direction of movement of the EA along the Y axis and continue vibrating the substrate (as described above) to generate (i) an additional line of additional filament and (ii) an additional one or more paddles, as described above.

[0044] In some embodiments, the processor is configured to control the motion assembly to repeat the generation of the filaments between about 30 and 100 times to generate a bundle of filaments. It is noted that the filaments are tightly bundled by (i) using a paddle as a joining hub for two or more filaments of the bundle, and (ii) intertwining the filaments of the resulting bundle by vibrating the substrate along the X-axis. In some embodiments, the processor is configured to control the EA and the motion assembly to generate paddles at random locations within the region. Furthermore, at least one of the paddles, typically all of the paddles, are positioned to join a random number of filaments. For example, a first paddle may join about 10 filaments, a second paddle may join about 50 filaments, and a third paddle may join about 70 filaments. The formation of the filaments and bundles is described in detail below in FIG. 1.

[0045] In other embodiments, the size and position of the paddle relative to the filaments formed on the surface of the substrate dictates the number of filaments that are joined by the paddle.

[0046] In some embodiments, the motion assembly is further configured to rotate the substrate about the axis of rotation to generate a layer of fabric or garment. A layer may be generated using a crisscross of two or more arrays of bundles. In such an embodiment, the processor is configured to control the motion assembly to (i) pause the movement of the EA and the vibration of the substrate, (ii) rotate the substrate, for example, at an angle of about 90 degrees (or any other suitable angle), and (iii) resume the vibration of the substrate and the movement of the EA along the first axis while discharging the molten granules onto the substrate.

[0047] Additionally or alternatively, the processor is configured to control the motion assembly to perform rotations at any angle other than 90 degrees. Embodiments relating to crisscrossing and layer formation are described in detail in Figures 3, 4, 5 and 6 below.

[0048] In some embodiments, the processor is configured to determine the number of layers and the construction of each layer for each of the garments.

[0049] In some embodiments, the mechanical structure of each layer (e.g., the number of bundles, the number of filaments and / or paddles per bundle, and the crisscross angle between the array of bundles arranged in the mesh) can determine the physical properties of the respective layer.

[0050] In some embodiments, the processor is configured to implement various techniques, such as machine learning, to determine the mechanical structure based on the layers and desired physical properties of the garment. It is noted that each garment is produced using a software-based recipe that defines the physical properties of the garment, thus improving the versatility and flexibility of the system to switch between producing first and second items by extracting the respective recipes from the system's memory. An embodiment relating to such techniques is described in more detail in FIG. 7 below.

[0051] In some embodiments, the processor is configured to control the system to (i) produce a front and a back of the garment, (ii) apply a cushioning sheet between the front and back, the cushioning sheet being configured to prevent undesired bonding between the front and back, (iii) bond between the edges of the front and back, for example by sewing or stitching, and (iv) remove the cushioning sheet to finish production of the garment. The process of manufacturing a garment by sewing the front and back is described in detail in FIG. 8 below.

[0052] The disclosed technology improves the flexibility, cost-effectiveness, customizability, and environmental friendliness of fabric and garment manufacturing processes.

[0053] System Description

[0054] FIG. 1 is a schematic diagram of a system 11 for manufacturing fabrics and / or garments, according to one embodiment of the present invention.

[0055] In some embodiments, system 11 includes an extrusion assembly (EA) 77, a motion assembly 13, and an operator console 65 having a processor 66, a memory 75, input devices 69 (such as, but not limited to, a mouse and keyboard), and a display 70 for displaying (i) images 71 of fabrics and / or garments being processed by system 11, and / or (ii) any other suitable information related to system 11 and the fabric and / or garment manufacturing process.

[0056] Typically, processor 66 of console 65 includes a general-purpose processor that is programmed with software to carry out the functions described herein. The software may be downloaded to processor 66 in electronic form, for example over a network, or alternatively or additionally may be provided and / or stored on a non-transitory tangible medium, such as magnetic, optical, or electronic memory (e.g., memory 75).

[0057] In some embodiments, the console 65 includes a driver 68 configured to drive the EA 77, described below, and some of the motors of the motion assembly 13. The system 11 includes a cable 74 configured to exchange signals between the console 65 and (i) the EA 77, and (ii) the motion assembly 13.

[0058] In some embodiments, the EA77, whose structure and operation are described in detail in Figures 2A and 2B below, is configured to receive granules of one or more substances produced from recycled clothing or any other source. In the example of Figure 1, the EA77 is configured to melt the granules and project the molten granules onto a surface of the movable substrate 44 (in a process described in Figures 2A and 2B below).

[0059] In some embodiments, substrate 44 is fabricated from a suitable metal (e.g., steel, stainless steel, or aluminum) and a magnetic film (not shown) is disposed on substrate 44. After the fabric or garment is fabricated, the magnetic film is removed from substrate 44 and then the finished fabric or garment is removed from the magnetic film.

[0060] In other embodiments, the substrate 44 is made from any suitable material and is coated with a film of polytetrafluoroethylene (PTFE), also known as Teflon, so that the filaments 22 are attached to the surface of the substrate 44, but the completed fabric can be removed from the substrate 44.

[0061] In some embodiments, the motion assembly 13, which may be implemented using any suitable type of translational and rotational stage, is configured to move the EA 77 and to move and / or oscillate the substrate 44 at least in the XY plane of an XYZ coordinate system. In the example of Figure 1, the motion assembly 13 is configured to move the EA 77 relative to the substrate 44 along the Y axis of the XYZ coordinate system. The motion assembly 13 is configured to oscillate the substrate 44 in a direction 12 (indicated by a double-headed arrow) that has at least a component parallel to the X axis of the XYZ coordinate system, or a direction that is a combination of the X and Y axes.

[0062] In some embodiments, a combined motion of the EA 77 and substrate 44 in two different axes causes the EA 77 to generate a line of filament 22. The line is generated by melting (fully or partially) granules within the EA 77 in order to extrude the molten granules and attach them to the substrate 44.

[0063] In some embodiments, the motion assembly 13 is configured to vibrate the substrate 44 using any suitable frequency or combination of frequencies defined by the processor 66 and implemented using a waveform signal generated by the driver 68.

[0064] In some embodiments, the EA 77 is configured to generate one or more puddles 33 of molten granules along the line of the filament 22. Embodiments relating to techniques for generating the puddles 33 are described below in Figures 2A and 2B.

[0065] In some embodiments, the EA77 is configured to generate section 9 of the line of filament 22 by drawing a portion of the molten granules from the paddle 33 and depositing the drawn portion on the substrate along the Y-axis (or along the curve of section 9 shown in the example of FIG. 1). More specifically, the EA77 is configured to draw a portion of the molten granules from the paddle 33 while moving (by the motion assembly 13) toward the intended position of the paddle 33a. The EA77 is then configured to generate the paddle 33a and then continue to move by the motion assembly 13 in a direction 10 that is typically parallel to the Y-axis, as shown in the example of FIG. 1. The EA77 is configured to draw a portion of the molten granules from the paddle 33a while moving in the direction 10 from the paddle 33a to generate the next line of filament 22.

[0066] In some embodiments, the processor 66 is configured to control the EA 77 and the motion assembly 13 to position the filament 22 and one or more paddles (e.g., paddles 33 and 33a) in a predetermined region of the substrate. In the context of the present disclosure and claims, the term "region" refers to a curve (e.g., of section 9) that is disposed along a region rather than a single axis due to the combination of the movement of the EA 77 in direction 10 (e.g., along the Y axis) and the vibration of the substrate 44 in direction 12, as described above. In the context of the present disclosure and claims, (i) the movement of the EA 77 in direction 10 and (ii) the vibration of the substrate 44 in direction 12 are performed using first and second movement profiles, respectively, which may be defined using one or more motion algorithms executed by the processor 66 or by any other suitable processing device configured to receive suitable software-implemented algorithms and apply the algorithms to the motion assembly 13.

[0067] In some embodiments, after generating the line of filament 22, processor 66 is configured to control motion assembly 13 to reverse the direction of movement of EA 77, e.g., along the Y-axis opposite direction 10. At the same time, processor 66 is configured to control motion assembly 13 to continue vibrating substrate 44 to generate (i) additional lines of additional filament 22, and (ii) one or more additional paddles 33, as described above. In the example of FIG. 1, paddles 33b and 33d are generated while EA 77 is moving in direction 10, and paddle 33c is generated while EA 77 is moving in a direction opposite direction 10.

[0068] In some embodiments, the processor 66 is configured to control the motion assembly 13 to repeat the generation of the filaments 22 (using the techniques described above) between about 30 and 100 times (or using any other suitable number of iterations) to generate a bundle 55 made from a plurality of filaments 22. In the context of this disclosure and in the claims, the term bundle also refers to a fiber made from a single filament 22, or from two or more filaments 22 entangled with one another using any suitable technique of the present disclosure.

[0069] In some embodiments, the filaments 22 are tightly bundled by (i) using a paddle 33 as a joining hub or anchor for two or more filaments 22 of the bundle, and (ii) intertwining the filaments 22 of the resulting bundle by vibrating the substrate 44 in the direction 12 (e.g., along the X-axis) while moving the EA 77 back and forth along the Y-axis.

[0070] In some embodiments, the motion assembly 13 is further configured to rotate the substrate 44 clockwise and counterclockwise in the direction 14 (indicated by a double-headed arrow). Rotation of the substrate 44 may occur about an axis of rotation (e.g., parallel to the Z-axis of an XYZ coordinate system) to generate layers of fabric or clothing. Embodiments relating to layering combinations are described in detail below in Figures 3, 4, 5, and 6.

[0071] In the context of this disclosure and the claims, the term "about" or "approximately" in relation to any numerical value or range indicates an appropriate dimensional tolerance that enables the collection of parts or components to function for its intended purpose as described herein.

[0072] This particular configuration of system 11 and the process for producing the filaments and bundles is simplified and presented as an example to explain the particular problems addressed by embodiments of the invention and to show the application of these embodiments in improving the performance of such systems, however, embodiments of the invention are in no way limited to this particular type of example system, and the principles described herein may be applied to other types of systems as well.

[0073] 2A is a schematic front view of the EA77 of FIG. 1 above according to an embodiment of the present invention, and FIG. 2B is a schematic side view of the EA77 of FIG. 1 above according to an embodiment of the present invention.

[0074] Now, reference is made to Figure 2A.

[0075] In some embodiments, the EA77 has a funnel-shaped feeder 19 configured to receive and direct the granules 20 into a column 34 of the EA77.

[0076] In some embodiments, the granules 20 may be manufactured from recycled textiles and clothing, or from any other source. The granules 20 may include meltable polymers, such as thermoplastics or thermo-soft plastics, which are plastic polymeric materials that become pliable or moldable at a certain elevated temperature, or melt completely, and then solidify upon cooling. For example, the granules 20 may include (i) polyester, (ii) nylon, (iii) polypropylene, and (iv) any suitable blend thereof blended with cotton, such that when melted, the molten granules include a mixture and / or compound of thermo-soft plastic and cotton.

[0077] In some embodiments, the processor 66 is configured to define materials and mixtures thereof for each fabric or garment. For example, a first fabric may include about 90% cotton and about 10% thermoplastic and other additives (e.g., for coloring the fabric), and a second fabric may include about 60% cotton and about 40% thermoplastic and other additives.

[0078] The size of the thermosoft plastic granules 20 may range between about 1 μm to about 10 mm, and may include a uniform size or a distribution of sizes of granules 20 within one or more batches of granules inserted into the feeder 19.

[0079] In some embodiments, the EA77 is configured to move the granules 20 along the column 34 towards a nozzle 27 located at the end of the column 34. The EA77 has one or more screw extruders 32 configured to rotate about a Z-axis using an electric motor 21 controlled by a processor 66. In this example, the EA77 is integrated within the interior volume of the column 34 and includes screw extruders 32a and 32b configured to move the granules 20 along the Z-axis within the column towards the substrate 44. In other embodiments, the EA77 may have a single screw extruder.

[0080] In some embodiments, the EA 77 has a first heating unit 16 configured to heat section 15 of the column 34, and a second heating unit 18 configured to heat section 17 of the column 34. Both heating units 16 and 18 are controlled by the processor 66 to set the temperature of sections 15 and 17, respectively, based on the electrical resistance of the heating elements (or using any other suitable technique). In this example, the heating unit 16 is configured to increase the temperature of section 15 to about 150°C, which is above the glass transition temperature of the granules 20 and increases the viscosity and flexibility of the granules 20. Furthermore, the heating unit 18 is configured to increase the temperature in section 17 to about 180°C, which is above the melting temperature of the granules 20. As shown in Figures 2A and 2B, the granules 20 are flexible in section 15 and melt to produce a molten polymer that may or may not have particles (e.g., cotton) immersed in the molten polymer formed by melting the granules.

[0081] In some embodiments, the EA 77 has cooling ribs 26 configured to dissipate heat generated in sections 15 and 17 to maintain the structure of the granules 20 when inserted into the feeder 19.

[0082] In some embodiments, the EA 77 is configured to dispense molten granules onto a surface of the substrate 44 via the nozzle 27. During dispensing of the molten granules, the processor 66 controls the motion assembly 13 to move the EA 77 along the Y axis and oscillate the substrate in the direction 12, as described in FIG. 1 above. In such embodiments, the EA 77 is configured to generate a line of the filament 22 on the substrate 44.

[0083] Reference is now made to inset 28. In another embodiment, instead of nozzle 27, EA 77 has a nozzle assembly 29 having multiple nozzles 27, each of which is configured to deposit molten granules onto the surface of substrate 44 to simultaneously produce multiple lines of respective filaments (to increase fabric production rate).

[0084] Referring now to the overall view of Figure 2A, in some embodiments, the EA 77 has a shaft 24 configured to transmit rotation generated by a motor 21 to the screw extruders 32a and 32b to move the granules 20. The EA 77 further has a spring 25 and a motor 23 configured to compress the spring by moving the bracket 7 towards the motor 21.

[0085] Reference is now made to FIG. 2B. In some embodiments, when the spring 25 is in a compressed position and the processor 66 controls the motor 23 to release the bracket 7, the spring is released and pushes the molten granules towards the substrate 44. In such embodiments, the kinetic energy of the spring 25 is transferred to the molten granules in the column 34 to increase the flow rate of the molten granules towards the substrate 44. As the moving speed of the EA 77 is constant, the increased flow rate of the molten granules results in the formation of a paddle 33 on the surface of the substrate 44 as described in FIG. 1 above. It should be noted that in practice, the shaft 24 is generally not shortened and the motor 21 is not moved towards the column 34 as shown in FIG. 2B, but the drawing of FIG. 2B is used for conceptual clarity to illustrate one embodiment related to the formation of the paddle 33. Furthermore, the motor 23 and the cooling ribs 26 are intentionally omitted from FIG. 2B for conceptual clarity.

[0086] In other embodiments, instead of or in addition to the spring 25 and motor 23, the EA 77 may have any other suitable mechanism for increasing the flow rate of molten granules during a predetermined time interval to create a puddle 33 on the surface of the substrate 44.

[0087] In some embodiments, the EA 77 has an air curtain 37 configured to improve the shaping and attachment of the filaments 22 and paddles 33 at their intended locations on the surface of the substrate 44. Note that the air curtain 37 is only shown in FIG. 2B for simplicity and clarity of presentation, and typically surrounds at least a portion, and typically all sides, of the column 34.

[0088] In other embodiments, the EA 77 may have one or more additional feeders (not shown) configured to supply one or more respective colors to the columns 34 to control the color of the filaments 22 and paddles 33.

[0089] In some embodiments, at least some of the granules 20 may have one or more colors to produce any suitable color of filament 22. Additionally or alternatively, one or more colors may be (i) added to the column 34 using, for example, a liquid feeder (not shown) and / or (ii) applied to one or more of the filaments 22 and / or bundles 33 after they are formed on the substrate 44.

[0090] This particular configuration of the EA77 is simplified and shown as an example to illustrate the particular problems addressed by aspects of the present invention and to show the application of these embodiments in improving the performance of such systems, however, embodiments of the present invention are in no way limited to this particular type of exemplary assembly and system, and the principles described herein may be applied to other types of liquid dispensers and other types of systems for producing garments and fabrics as well.

[0091] Fabric and clothing layer generation

[0092] Figures 3, 4, 5 and 6 are schematic diagrams of arrangements of bundles 55 in respective fabric layers 40, according to embodiments of the present invention. In some embodiments, each of the garments or fabrics includes one or more layers, and the bundles of Figures 3-6 are produced using the techniques described in Figures 1, 2A and 2B above.

[0093] In some embodiments, the motion assembly 13 is configured to rotate the substrate 44 clockwise and counterclockwise about an axis of rotation in the direction 14 (shown in FIG. 1 above) to generate each fabric or garment layer 40.

[0094] Reference is now made to FIG. 3, which shows fabric layer 40a.

[0095] In some embodiments, layer 40a includes four arrays of bundles 55a, 55b, 55c, and 55d arranged in a crisscross configuration of meshes and is generated using the process sequence described herein. In one example of a process sequence, processor 66 controls EA 77 and motion assembly 13 to (i) generate an array of bundles 55a, (ii) rotate substrate 44 clockwise through an angle of about 90 degrees, (iii) generate an array of bundles 55b, (iv) rotate substrate 44 counterclockwise through an angle of about 90 degrees, (v) generate an array of bundles 55c, (vi) rotate substrate 44 clockwise through an angle of about 90 degrees, and (vii) generate an array of bundles 55d. It should be noted that the process sequence described above allows for the intertwining of the bundles of layer 40a. Additionally, processor 66 controls motion assembly 13 to suspend vibration while rotating substrate 44 clockwise or counterclockwise.

[0096] 3, bundles 55a and 55c are positioned on substrate 44 so that they are orthogonal to bundles 55b and 55d and also orthogonal to the edge of layer 40a. The orthogonality between the bundles is achieved by the rotation angle of substrate 44, and the orthogonality between the bundles and the edge of layer 40a is achieved in the setup of system 11.

[0097] Reference is now made to Figure 5, which illustrates the structure of layer 40c. In some embodiments, bundles 55a and 55c are positioned on substrate 44 so that they are orthogonal to bundles 55b and 55d, but not orthogonal to the edges of layer 40c. For example, angle 42a between bundles 55c and 55d is approximately 90 degrees, and angle 42b between bundles 55c is less than 90 degrees.

[0098] 6, which illustrates the structure of layer 40d. In some embodiments, bundles 55a and 55c are not orthogonal to bundles 55b and 55d, and are not orthogonal to the edge of layer 40d. For example, angle 42c between bundles 55a and 55b is less than 90 degrees, and angle 42d between bundle 55a and the edge of layer 40d is also less than 90 degrees.

[0099] Reference is now made to Figure 4, which illustrates the structure of layer 40b. In some embodiments, layer 40b has one or more reinforcement points (RP) 88 created by welding between interlocking bundles 55, for example by locally heating bundle 55c, which is placed above bundle 55b. In other embodiments, RP 88 may be created using a fusion process.

[0100] In the examples of Figures 40b and 40d, all the RP88 are similar, but their density and arrangement will affect the mechanical properties of each layer, such as flexibility and tensile strength. Note that the RP88 are uniformly arranged in Figures 4 and 6.

[0101] In other embodiments, the distribution of RP88 may not be uniform along the layer, for example a first section of the layer may have RP88 and another section of the layer may not have RP88 or may have a lower density of RP88 compared to the first section.

[0102] Referring back now to FIG. 3, in some embodiments, layer 40a may have different types of RP88, e.g., RP88a in a first section and RP88b different from RP88a in a second section. The difference between RP88a and RP88b may be obtained by the amount and type of fusion material and / or in the fusion or welding process applied to each type of RP88. The difference between RP88a and RP88b may result in, for example, different tensile strengths in the respective sections.

[0103] In other embodiments, the arrangement of bundles 55 and garment specifications may not require the use of RP88 in one or more layers of each garment. For example, layer 40c of FIG. 5 may not have any RP88. Additionally, a given garment may have multiple layers that may be similar to each other or different from each other. For example, a given garment may have an inner layer 40c (which does not have RP88 and is therefore smoother) and an outer layer 40a that is more rigid.

[0104] In some embodiments, the processor 66 is configured to determine the number of layers 40 in each garment and the structure of each layer within the garment (e.g., the number of filaments 22 and paddles 33 within each bundle 55, and the density and arrangement of the bundles 55).

[0105] In some embodiments, the density of the bundles 55, their crisscross arrangement, the type and density of RP88, and the number and arrangement of layers may result in different physical properties of the final garment. For example, tensile strength, ductility (e.g., level of stretch), permeability to light and / or air, reflectivity, weight, breathability, foldability, and durability may all be affected by the construction of one or more layers of the garment's fabric. Additionally, each layer of the garment and / or different sections of the fabric may have different constructions that may result in different physical properties.

[0106] In other embodiments, instead of a mesh arranged in a predetermined crisscross configuration, processor 66 is configured to control EA 77 and motion assembly 13 to arrange one or more bundles 55 in an irregular pattern arranged in a suitable two-dimensional (2D) or three-dimensional (3D) configuration. In such embodiments, processor 66 is configured to control EA 77 to apply RP 88 to selected locations of the pattern to control one or more physical properties of the layers and / or fabric produced by system 11. Additionally or alternatively, processor 66 is configured to control the laser assembly to direct a laser beam to predetermined locations of the pattern to weld two or more bundles 55 together.

[0107] In an alternative embodiment, the processor 66 is configured to control the EA 77 and the motion assembly 13 to arrange one or more bundles 55 using a plurality of layers 40 arranged in a three-dimensional configuration, where each layer 40 may have any suitable shape other than crisscrossing lines. For example, a first section of the first layer 40y may have one or more bundles 55 arranged in a set of concentric or non-concentric circles, and a second section of the second layer 40z may have any suitable arrangement of bundles 55 that may be produced on the first section of the first layer 40y and welded to one or more sections of bundles 55 arranged in a set of concentric or non-concentric circles. Such non-crisscrossing structures may define physical properties in one or more respective sections of a respective garment that are not obtainable using the crisscrossing configuration described in Figures 3-6 above.

[0108] The structures of layers 40a-d and layers 40y-z are provided as examples, and the principles described herein may be similarly applied to other types of layers and one or more sections of other layers of fabrics or garments produced by system 11 using the techniques described above.

[0109] FIG. 7 is a block diagram that illustrates generally a machine learning based module 50 according to an embodiment of the present invention, which is configured to automatically determine the mechanical structure of a fabric based on its physical properties.

[0110] In some embodiments, module 50 may be implemented in software in processor 66 or in hardware using any suitable device. Module 50 has a machine learning (ML) engine 54 that is trained (as described herein) to receive inputs 52 of attributes of physical properties of a given fabric or garment and output a recipe 56 for system 11 that includes the mechanical structure of filaments 22, paddles 33, bundles 55 and layers 40 and the process flow for producing them.

[0111] In some embodiments, the ML engine 54 may be implemented using any suitable type of artificial neural network. Artificial neural networks (ANNs) are statistical models directly inspired and partially modeled on biological neural networks. They can model and process nonlinear relationships between inputs and outputs in parallel. In this example, the ML engine may include one or more neural networks, such as a convolutional neural network (CNN), or any other suitable type. A CNN includes (i) one or more dropout layers, (ii) multi-layer convolution filters implemented with appropriate tensors of convolutional layers, and (iii) one or more fully connected layers, also referred to herein as dense layers.

[0112] In some embodiments, the processor 66 is configured to receive information about existing fabrics of interest, which may include, but are not limited to, physical properties such as the fabric's or garment's tensile strength, ductility (e.g., elongation level), permeability to light and / or air, reflectivity, smoothness, weight, breathability, color(s), foldability, and durability.

[0113] Additionally or alternatively, the processor 66 is configured to receive information related to the corresponding mechanical structure of the fabric in question. For example, this information may include surface characteristics, bundle thickness, color and bundle pattern, and bundle arrangement (e.g., as described in Figures 3-6 above), spacing between (i) adjacent rows and (ii) adjacent columns of bundles, bundle orientation, number of layers, number of filaments and paddles in each type of bundle, and any other suitable information related to the structure of the fabric in question.

[0114] In some embodiments, the processor 66 is configured to store a database in the memory 75 that includes the physical properties and corresponding structures of each layer and each fabric or garment. The processor 66 may use this database to train the ML engine 54, for example, using supervised learning or reinforcement learning techniques.

[0115] In some embodiments, multiple types of training sets may be used to train the CNN of the ML engine 54. For example, (i) each fabric and garment may be classified based on any suitable classification, such as the physical properties described above, (ii) a camera may be used at an appropriate magnification to acquire images of the available fabrics and garments of each class, (iii) for each fabric or garment, (a) one or more images and (b) an input set of labels indicating the physical properties are inserted as inputs into the CNN, (iv) the input images are operated on by dropout layers, convolutional layers, and dense layers, and (v) the CNN outputs the structure of each layer of the fabric or garment intended to be produced. In this example, the output may include at least one parameter selected from a list of parameters consisting of: (i) a size of each bundle 55 (e.g., a thickness defined by the number of filaments 22); (ii) a first distance between adjacent bundles 55 along a first axis of the fabric or garment; (iii) a second distance between adjacent bundles 55 along a second axis of the fabric or garment; (iv) an angle 42 between each pair of bundles 55 disposed along the first axis and the second axis; (v) an estimated location of reinforcement points (RPs) 88 in the lattice of bundles 55 disposed along the first axis and the second axis; and any other suitable parameter.

[0116] In some embodiments, the processor 66 is configured to use supervised learning techniques to train the CNN by feeding back corresponding parameters, such that after a certain number of training sets, the CNN model converges for at least one class of fabric and clothing. Additionally or alternatively, reinforcement learning may be used to train the CNN. For example, after receiving output parameters from the CNN, a user or the processor 66 may provide the CNN with (i) OK or not OK feedback for the entire set of output parameters or each output parameter, or (ii) an estimated accuracy of the output (e.g., in percentages). Each output parameter may have a weight and a score, and the processor 66 is configured to calculate an estimated accuracy based on the weight and score of each output parameter. The CNN is configured to converge after a suitable number of training sets.

[0117] In some embodiments, processor 66 is configured to maintain one or more thresholds indicative of an accuracy criterion for the CNN, and once the accuracy criterion is met for one or more validation sets, the CNN has a trained model and training of ML engine 54 is terminated.

[0118] In some embodiments, the processor 66 is configured to generate a user interface (UI) for the fabric and / or garment designer. The UI may include input fields for attributes of the physical properties of the designed fabric or garment, and the processor 66 is configured to output the layers and / or fabric or garment structure using the trained ML engine 54 or any other suitable artificial intelligence (AI) techniques.

[0119] In some embodiments, the processor 66 is configured to output a model of the fabric or garment and the corresponding structure of each of its layers. In some embodiments, after validation and / or optimization of the inputs from the user, the processor 66 is configured to generate one or more recipes that can be used by the system 11 to generate the desired fabric or garment. At least one of the recipes can be generated during the inference stage of the trained ML engine 54.

[0120] FIG. 8 is a schematic diagram of a shirt 61 produced using system 11, according to one embodiment of the present invention.

[0121] Typically, shirts 61 and other types of garments have a front and a back. In some embodiments, processor 66 is configured to control EA 77 and motion assembly 13 to generate the front and back of shirt 61 using the techniques described in Figures 1-7 above and Figure 9 below.

[0122] In some embodiments, each face of shirt 61 may be generated as a single polygon (e.g., a rectangle), and then redundant sections 64 of the polygon are removed (e.g., by cutting them) to generate the design shape of shirt 61. As described above in FIG. 2A, sections 64 may be recycled to generate granules 20 that can be used to generate additional garments, such as shirt 61.

[0123] In this example, the processor 66 is configured to control the EA 77 and the motion assembly 13 to generate the design shape directly, so that redundant fabric (e.g., fabric section 64) is not removed. This technique reduces waste and / or recycled resources, thus improving the environmental friendliness of the production process.

[0124] In some embodiments, after producing the front and back of shirt 61, sheet 62 (also referred to herein as a release layer) is placed between the front and back of shirt 61 to prevent undesirable bonding between the front and back.

[0125] In some embodiments, sheet 62 comprises any suitable material, such as an acrylonitrile butadiene styrene (ABS) polymer with polypropylene configured to prevent bonding between the front and back surfaces (also referred to herein as peeling between the front and back surfaces of shirt 61.) In other embodiments, any suitable technique other than a release layer may be used to prevent bonding.

[0126] In some embodiments, sheet 62 has the designed shape of shirt 61, but the size of sheet 62 is slightly smaller than the size of shirt 61 so as to align front and back edges 67 of shirt 61 and place them in direct contact with one another. In such embodiments, the front and back edges 67 of shirt 61 are bonded together using any suitable technique. For example, a laser beam may be used to heat the edges 67 and bond them together, or a stitching machine may be used to sew the edges together, or any other suitable technique may be used to bond between the front and back edges 67 of shirt 61.

[0127] In another embodiment, a sheet 65 having a polygonal shape and made from a suitable release material as described for sheet 62 above may be placed between the front and back of shirt 61 and folded before bonding between the respective edges 67 of the front and back of shirt 61.

[0128] Figure 9 is a flow chart that illustrates a schematic of a method of manufacturing a garment according to one embodiment of the present invention. In this example, the garment comprises the shirt 61 of Figure 8 above.

[0129] The method begins with a definition step 100 using a processor 66 or any other suitable device configured to define process parameters for obtaining a selected mechanical structure for each layer of shirt 61. At least one of the layers may have one or more sections, each of the sections having a different structure of filaments, bundles, and other components. The components and process parameters are described in detail in Figures 1-8 above. More specifically, based on a list of desired physical properties, the mechanical structure of shirt 61 may be defined using machine learning (ML) techniques described in Figure 7 above.

[0130] In a granule insertion step 102, a user or robot of the system 11 inserts granules 20 into the feeder 19 of the extrusion assembly (EA) 77, as described above in Figures 2A and 2B.

[0131] In the granule movement step 104, the EA77 is configured to move the granules 20 into a column 34 of the EA77, after which one of more screw extruders 32 of the EA77 is rotated (e.g., by an electric motor 21) about the Z axis to move the granules 20 toward one or more nozzles 27, as described above in Figures 2A and 2B.

[0132] In the melting step 106, the processor 66 controls the heating units 16 and 18 to raise the temperature of the granules 20 in the respective sections 15 and 17 of the column 34 above the glass temperature (Tg) and above the melting temperature (Tm), as described above in FIG. 2A.

[0133] In a filament generation step 108, the processor 66 controls the EA 77 and the motion assembly 13 to generate the filament 22 and paddle 33 by depositing molten granules on the substrate 44 while moving the EA 77 relative to the substrate 44, and, optionally, by vibrating the substrate 44 in the direction 12, as described above in FIGS. 1, 2A, and 2B.

[0134] In a bundle and layer generation step 110, the processor 66 controls the EA 77 and the motion assembly 13 to generate bundles 55 and layers 40 with the selected crisscross configuration by (i) vibrating the substrate 44 to intertwine groups of filaments 22 with one another, (ii) grouping multiple filaments 22 with the paddles 33, (iii) rotating the substrate 44 about the Z axis relative to the EA 77 to generate the desired crisscross configuration, and, optionally, (iv) welding or fusing one or more RPs 88 between selected bundles 33 that cross each other at predetermined locations within the respective layers 40, as described in detail in FIGS. 1-6. Note that steps 108 and 110 may be performed uniformly across the entire shirt 61, for example as described in FIG. 3 above, or may be performed differently in different sections of the garment or fabric.

[0135] In some embodiments, step 110 completes the manufacturing of first layer 40 of shirt 61 .

[0136] In decision step 112, processor 66 determines whether shirt 61 has additional layers 40. If additional layers 40 must be generated, the method loops back to step 100. If additional layers 40 do not need to be produced, the method proceeds to surface bonding step 114, which ends the method. In step 114, a front and back surface of shirt 61 are generated and front and back edges 67 of shirt 61 are bonded together to produce shirt 61, as described in detail in FIG. 8 below.

[0137] The method of FIG. 9 is simplified for conceptual clarity, and the process of producing shirt 61 may include additional steps such as, but not limited to, (i) the insertion of color (e.g., in column 34) to impart a desired color to shirt 61 or each section thereof, and (ii) the production of granules 20, for example, by recycling old clothing and fabrics.

[0138] 10 is an image of an EA78 according to another embodiment of the present invention. The EA78 can, for example, replace the EA77 in FIGS. 1, 2A and 2B above.

[0139] In some embodiments, the EA 78 includes a motor 23 and a bracket 7 disposed between the cooling ribs 26 and the heating unit 16. In such embodiments, the processor 66 is configured to control the motor 23 to move the bracket 7 upward along the Z axis (e.g., toward the motor 21) to deposit an increased amount of molten granules via the nozzle 27, thereby generating a puddle 33 on the surface of the substrate 44.

[0140] In some embodiments, the EA 78 has a fan 99 configured to blow air on the cooling ribs 26 to dissipate heat generated by the heating units 16 and 18, as described in FIG. 2A above, in order to maintain the structure of the granules 20 when inserted into the feeder 19 (shown in FIGS. 2A and 2B above).

[0141] Although the embodiments described herein primarily address the production of garments and textiles, the methods and systems described herein may be used in other applications.

[0142] Thus, the above embodiments are given by way of example, and it is understood that the present invention is not limited to what has been particularly shown and described above. Rather, the scope of the present invention includes both combinations and subcombinations of the various features described above, as well as variations and modifications thereof that would occur to one skilled in the art upon reading the above description, but that are not disclosed in the prior art. Documents incorporated by reference into this patent application should be considered as an integral part of this application, but if any term is defined in these incorporated documents in a manner that is inconsistent with a definition expressly or impliedly made herein, then only the definition in this specification should be considered.

Claims

1. an extrusion assembly (EA) configured to melt granules of one or more substances and project the molten granules onto a substrate; a motion assembly configured to move the EA along a first axis relative to the substrate and oscillate the substrate along a second axis different from the first axis while the EA ejects the molten granules to generate a line of filaments on the substrate; and a processor configured to control the EA and the motion assembly to place the filament at a predetermined area of ​​the substrate. system.

2. The EA is configured to generate one or more puddles of the molten granules along the line. The system of claim 1 .

3. the EA is configured to generate a section of the line by drawing a portion of the molten granules from the puddle and depositing the drawn portion on the substrate along the first axis; The system of claim 2.

4. The processor is configured to: (i) determine the predetermined region by controlling a first movement profile for moving the EA along the first axis and a second movement profile for oscillating the substrate along the second axis; A system according to any one of claims 1 to 3.

5. the processor is configured to control the motion assembly to move the EA in an opposite direction along the first axis to place additional filaments at additional predetermined areas of the substrate. A system according to any one of claims 1 to 3.

6. At least a portion of the predetermined region overlaps with at least a portion of the additional predetermined region. The system of claim 5.

7. the processor is configured to control the motion assembly and the EA to generate a plurality of the filaments and the additional filaments and to generate a first array of one or more bundles by coupling between two or more of the filaments and the additional filaments in each of the bundles. The system of claim 6.

8. the processor is configured to control the motion assembly and the EA to generate at least one of the bundles by intertwining one or more of the filaments with one or more of the additional filaments. The system of claim 7.

9. the processor is configured to control the motion assembly and the EA to generate at least one of the bundles by bonding two or more of the filament and the additional filament to the puddle of molten granules. The system of claim 7.

10. the motion assembly is configured to rotate the substrate about an axis of rotation, and the processor is configured to (i) control the motion assembly to rotate the substrate through a predetermined rotation angle, and (ii) control the motion assembly and the EA to generate one or more second arrays of the bundles on the substrate. The system of claim 7.

11. the processor is configured to control the motion assembly and the EA to generate a layer including at least two of the bundles of the first and second arrays intertwined with each other. The system of claim 10.

12. the processor is configured to control the motion assembly and the EA to generate the entangled bundles in a crisscross arrangement. The system of claim 11.

13. the processor is configured to control the motion assembly and the EA to generate the entangled bundles in first and second orientations, respectively, the first and second orientations defining a given angle between the entangled bundles. The system of claim 12.

14. The processor is configured to control the motion assembly and the EA to generate one or more reinforcement points (RP) between one or more pairs of the intertwined bundles. The system of claim 11.

15. the processor is configured to control the motion assembly and the EA to generate at least one of the RPs by welding together at least one of the pairs; The system of claim 14.

16. The processor is configured to control the motion assembly and the EA to generate at least one of the RPs by blending between at least one of the bundles of the pair. The system of claim 14.

17. the processor is configured to control the motion assembly and the EA to: (i) generate one or more layers and generate a first side of a fabric by stacking a first plurality of the layers; and (ii) generate one or more additional layers and generate a second side of a fabric by stacking a second plurality of the additional layers. The system of claim 11.

18. the first surface having at least a first edge and the second surface having at least a second edge, and the processor is configured to control a bonding assembly to bond between the first edge and the second edge to generate the fabric.

20. The system of claim 17.

19. the processor is configured to control a robot to place a release layer between the first surface and the second surface to prevent undesired bonding between the first surface and the second surface.

20. The system of claim 18.

20. the first surface is disposed over the second surface such that the first edge and the second edge are aligned, and the processor is configured to control a laser source to direct a laser beam to the aligned first edge and the second edge for coupling between the first surface and the second surface.

20. The system of claim 18.

21. the first surface is positioned over the second surface such that the first edge and the second edge are aligned, and the processor is configured to control a stitcher to stitch between the first edge and the second edge.

20. The system of claim 18.

22. The granules have a size between 1 μm and 10 mm before being melted. A system according to any one of claims 1 to 3.

23. At least one of the granules is made from recycled textiles or recycled clothing. A system according to any one of claims 1 to 3.

24. At least one of the granules comprises a meltable polymer selected from the list of meltable polymers consisting of one or more of: (i) polyester, (ii) nylon, and (iii) polypropylene; A system according to any one of claims 1 to 3.

25. At least one of the granules comprises the meltable polymer mixed with cotton.

25. The system of claim 24.

26. The EA comprises a column and one or more screw extruders disposed within the column and configured to move the granules along the column. A system according to any one of claims 1 to 3.

27. The EA has one or more heating units configured to increase the temperature in a given section of the column to melt the granules.

27. The system of claim 26.

28. The one or more heating units include (i) a first heating unit disposed in a first section of the column and configured to heat the granules to a first temperature to increase flexibility of the granules, and (ii) a second heating unit disposed in a second section of the column different from the first section and configured to heat the granules to a second temperature higher than the first temperature to melt the granules.

28. The system of claim 27.

29. the EA having one or more cooling ribs configured to dissipate heat generated in the first section and the second section from heating a third section of the column; 30. The system of claim 28.

30. The EA has a motor configured to rotate at least one of the screw extruders to move the granules along the column.

27. The system of claim 26.

31. the EA has one or more nozzles, at least one of the nozzles configured to generate the lines by depositing the molten granules on the substrate; A system according to any one of claims 1 to 3.

32. the nozzles having (i) a first nozzle configured to generate a first line in a first region of the substrate; and (ii) a second nozzle configured to generate a second line in a second region of the substrate different from the first region.

32. The system of claim 31.

33. the EA has an air curtain configured to perform one or both of: (i) shaping the line of filaments produced on the substrate; and (ii) improving adhesion of the line of filaments to the substrate; A system according to any one of claims 1 to 3.

34. The EA has a mechanism for increasing the flow rate of the molten granules to generate a puddle along the line. A system according to any one of claims 1 to 3.

35. the mechanism includes a spring; 35. The system of claim 34.

36. a machine learning (ML) engine configured to receive an input including one or more attributes of one or more physical properties of a fabric and output a mechanical structure of the fabric; A system according to any one of claims 1 to 3.

37. The mechanical structure includes one or more of: (i) one or more of the filaments; (ii) one or more paddles; (iii) one or more bundles generated by intertwining a plurality of the filaments; and (iv) one or more layers including one or more arrays of first and second bundles intertwined with each other.

37. The system of claim 36.

38. The ML engine is configured to output a recipe for generating the mechanical structure within the system.

38. The system of claim 37.

39. the ML engine is implemented in software, and the processor is configured to execute the software; 37. The system of claim 36.

40. The fabric includes a first fabric section having a first physical property and a second fabric section having a second physical property different from the first fabric section and different from the first physical property, and the ML engine is configured to output a first mechanical structure of the first fabric section and a second mechanical structure of the second fabric section different from the first mechanical structure.

37. The system of claim 36.

41. The paddles are formed at random locations within the predetermined area.

38. The system of claim 37.

42. the paddles include (i) a first paddle for joining together a first number of the filaments; and (ii) a second paddle for joining together a second number of the filaments, the second number being different from the first number.

38. The system of claim 37.

43. Melting granules of one or more substances in an extrusion assembly (EA) and projecting the molten granules onto a substrate; and moving the EA relative to the substrate along a first axis and vibrating the substrate along a second axis different from the first axis while the EA expels the molten granules to generate a line of filaments on the substrate. method.

44. generating one or more puddles of the molten granules along the line; 44. The method of claim 43.

45. generating a section of the line by drawing a portion of the molten granules from the puddle and depositing the drawn portion on the substrate along the first axis.

45. The method of claim 44.

Images