Methods for preparing, applying, and controlling self-sensing artificial muscles

Abstract

This application discloses a method for preparing, applying, and controlling a self-sensing artificial muscle, relating to the field of flexible robotics. The method involves coating a stimulus-responsive soft material with a conductive coating; fabricating the coated soft material into conductive fiber bundles; and weaving the conductive fiber bundles into a knotted structure with deformation gaps to form a self-sensing artificial muscle. The stimulus-responsive soft material can deform under external stimuli, and the sensing signal can be transmitted through the conductive coating itself. The conductive coating can deform together with the driven stimulus-responsive soft material, thus eliminating the need for an external sensing unit and removing defects in size, wiring, and environmental adaptability. The deformation gaps in the knotted structure allow the conductive fiber bundles to slip relative to each other and redistribute strain during deformation, preventing conductive layer peeling or signal drift due to excessive local strain during long-term cyclic use; thereby achieving stable integration of driving and sensing functions.

Description

Technical Field

[0001] This application relates to the field of flexible robot technology, and in particular to a method for preparing, applying and controlling a self-sensing artificial muscle. Background Technology

[0002] Currently, in order to endow artificial muscles with self-sensing capabilities, it is common practice to physically integrate independent sensing units (such as strain gauges, flexible electrodes, optical fibers, etc.) with the drive unit, relying on these external sensing units to measure deformation or force.

[0003] However, external sensing units result in a bulky system with complex wiring, and are difficult to operate reliably in complex environments such as obstruction, underwater, or bending. Furthermore, the interface between the external sensor and the drive unit is unstable, and long-term use can easily lead to peeling or signal drift.

[0004] Therefore, the relevant technologies cannot achieve a stable integration of the driving and sensing functions of artificial muscles. Summary of the Invention

[0005] The main objective of this application is to provide a method for preparing, applying, and controlling a self-sensing artificial muscle, aiming to solve the technical problem of achieving stable integration of the driving and sensing functions of the artificial muscle.

[0006] To achieve the above objectives, this application proposes a method for preparing a self-sensing artificial muscle, the method comprising: A conductive coating is applied to the surface of a stimulus-responsive soft material; Soft materials coated with a conductive coating are made into conductive fiber bundles; The conductive fiber bundles are woven into a knotted structure with deformation gaps to form a self-sensing artificial muscle.

[0007] In one embodiment, the size of the deformation gap is determined based on the sensing sensitivity requirement.

[0008] In one embodiment, the knot structure is at least one of a square knot, a reverse knot, a figure-eight knot, a single knot, a lark's head knot, and a three-strand braid.

[0009] In one embodiment, the conductive fiber bundle includes a single strand of conductive fiber bundle. When the conductive fiber bundle is a single strand of conductive fiber bundle, the knot structure is a square knot. The step of weaving the conductive fiber bundle into a knot structure with a deformation gap includes: The single strand of conductive fiber is folded in half, and the two free ends formed after folding are tied into a flat knot, with a predetermined gap between the flat knot and the folding point, to form a knot structure with a deformation gap.

[0010] In one embodiment, the conductive fiber bundle includes a single-strand conductive fiber bundle and a multi-strand conductive fiber bundle, and the step of fabricating the soft material coated with a conductive coating into a conductive fiber bundle includes: A single strand of soft material coated with a conductive coating is used as a single-strand conductive fiber bundle. And / or, multiple strands of soft material coated with conductive coating are combined into a multi-strand conductive fiber bundle by arranging them in parallel, twisting or braiding.

[0011] In one embodiment, the stimulus-responsive soft material is either a liquid crystal elastomer fiber or a hydrogel smart material fiber.

[0012] In one embodiment, the conductive coating material includes at least one of stainless steel soft material, conductive polymer, carbon nanomaterial and metal nanowire.

[0013] This application also proposes a method for using a self-sensing artificial muscle, wherein the self-sensing artificial muscle is prepared by any of the above-described methods, and the self-sensing artificial muscle is used as a flexible actuator, an actuation component of a soft robot, or a power and sensing element in a wearable device.

[0014] This application also proposes a method for controlling a self-sensing artificial muscle, wherein the self-sensing artificial muscle is prepared using any of the self-sensing artificial muscle preparation methods described above, and the method for controlling the self-sensing artificial muscle includes: In response to the driving command, electrical signals are applied to both ends of the self-sensing artificial muscle, and the resistance values ​​at both ends of the self-sensing artificial muscle are collected synchronously. Based on the resistance value and a preset relationship model between the change in muscle length of the artificial muscle and the relative rate of change in resistance, the current muscle length of the self-sensing artificial muscle is determined. The electrical signal is adjusted based on the current muscle length to control the deformation of the self-sensing artificial muscle.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: This application involves coating a conductive coating onto the surface of a stimulus-responsive soft material; fabricating the soft material coated with the conductive coating into conductive fiber bundles; and weaving the conductive fiber bundles into a knotted structure with deformation gaps to form a self-sensing artificial muscle. It is understood that the conductive coating is directly coated onto the surface of the stimulus-responsive soft material, which can deform under external stimuli. The sensing signal can be transmitted through the conductive coating itself. Even in obstructed, liquid-covered, or bending environments, the conductive coating can still deform together with the driven stimulus-responsive soft material. Therefore, no external sensing unit is required, eliminating defects in size, wiring, and environmental adaptability. The deformation gaps in the knotted structure allow the conductive fiber bundles to slip relative to each other and redistribute strain during deformation, alleviating stress concentration at the interface between the conductive coating and the soft material. This avoids conductive layer peeling or signal drift caused by excessive local strain during long-term cyclic use, thus achieving stable integration of driving and sensing functions. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of the method for preparing self-sensing artificial muscle according to Example 1 of this application; Figure 2 This is a schematic diagram of the sensing principle of the self-sensing artificial muscle provided in Example 1 of the preparation method of the self-sensing artificial muscle of this application. Figure 3 This is a schematic diagram illustrating the changes in muscle length and resistance during the driving process provided in Embodiment 3 of this application.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Based on this, embodiments of this application provide a method for preparing a self-sensing artificial muscle, referring to... Figure 1 , Figure 1 This is a schematic flowchart of the first embodiment of the method for preparing self-sensing artificial muscle according to this application.

[0023] In this embodiment, the method for preparing the self-sensing artificial muscle includes steps S10 to S30: Step S10: Coat the surface of the stimulus-responsive soft material with a conductive coating; Step S20: The soft material coated with a conductive coating is made into a conductive fiber bundle; Step S30: The conductive fiber bundle is woven into a knot structure with deformation gaps to form a self-sensing artificial muscle.

[0024] It should be noted that stimulus-responsive soft materials refer to flexible materials that can undergo reversible changes in shape, volume, or modulus under the influence of external physical or chemical stimuli (such as heat, light, electricity, pH, humidity, ionic strength, etc.).

[0025] In one feasible embodiment, the stimulus-responsive soft material is either a liquid crystal elastomer fiber or a hydrogel smart material fiber. Liquid crystal elastomer fibers possess thermally induced phase change characteristics, enabling them to generate rapid and significant reversible contractile strain (up to 40%-60%) under Joule heating. They also exhibit high mechanical output strength, good cycle stability, and can operate stably without relying on specific humidity or chemical environments. In contrast, hydrogel smart material fibers typically rely on water molecule diffusion or pH changes for propagation, have a slower electrothermal response, and are prone to failure in dry environments. Therefore, this embodiment preferably uses liquid crystal elastomer fibers as the stimulus-responsive soft material.

[0026] Furthermore, the morphology of the stimulus-responsive soft material can be fibrous, membrane-like, or block-like. Since the fibrous morphology has a higher aspect ratio, it can provide greater deformation displacement and faster heat / mass transfer response speed in the same volume. In this embodiment, the morphology of the stimulus-responsive soft material is preferably fibrous.

[0027] By applying a continuous conductive coating to the surface of the prepared stimulus-responsive soft material using a pre-defined process, the stimulus-responsive soft material is transformed from an insulator or weak conductor into a composite structure with a conductive surface.

[0028] In one feasible embodiment, the conductive coating material includes at least one of stainless steel soft material, conductive polymer, carbon nanomaterials, and metal nanowires. Because stainless steel soft material possesses excellent conductivity, good mechanical flexibility, and fatigue resistance, it can maintain the structural integrity and electrical stability of the conductive coating during high-strain cyclic driving. Simultaneously, the stainless steel soft material exhibits good interfacial bonding with stimulus-responsive soft materials such as liquid crystal elastomer fibers, effectively mitigating the risk of conductive layer peeling or cracking caused by repeated deformation. Furthermore, stainless steel soft material is relatively inexpensive and easy to process (e.g., drawing, weaving, sputtering), making it suitable for mass production. In this embodiment, the conductive coating material is preferably stainless steel soft material.

[0029] Specifically, the implementation method for coating the surface of a stimulus-responsive soft material with a conductive coating can be any one of the following: water vortex wire wrapping method, magnetron sputtering method, conductive wire spiral winding method, dip-coating method, air gun spraying method, and screen printing method; specifically: A conductive fiber layer can be coated onto the surface of the prepared stimulus-responsive soft material fiber using a water vortex method. Specifically, conductive fibers (such as chopped carbon fibers or short metal fibers) are delivered into the water, and a water vortex is activated. The conductive fibers are then uniformly wrapped around and attached to the fiber surface under the influence of the vortex, forming a conductive fiber layer. This conductive fiber layer has a fluffy appearance on the fiber surface, which facilitates the formation of abundant contact points during subsequent knot weaving, enhancing sensing sensitivity.

[0030] Alternatively, in a vacuum chamber, using a stimulus-responsive soft material fiber as a substrate, a metal target (such as gold, silver, or copper) is used for magnetron sputtering in an argon atmosphere to deposit metal particles onto the fiber surface, forming a dense conductive coating. This conductive coating is dense and uniform, has strong adhesion, and its thickness is controllable, without altering the mechanical properties of the fiber itself. It is suitable for applications requiring high consistency and durability of the conductive layer.

[0031] Alternatively, conductive wires (including metal filaments, metal-coated fibers, or intrinsically conductive polymer filaments) can be spirally wound onto the surface of stimulus-responsive soft material fibers. Specifically, the fiber can be used as the core wire, and the conductive wires can be uniformly wound around the outer periphery of the fiber at a certain pitch (e.g., 1-3 mm) to form a spring-like conductive layer structure. This conductive layer structure is independent, has high mechanical strength, can maintain the integrity of the conductive path under ultra-large strain conditions, and the winding pitch can be adjusted to adapt to different deformation ranges.

[0032] Using the above method, the conductive coating and the stimulus-responsive soft material form an integrated structure during the preparation stage. There is no post-bonding interface between the conductive coating and the stimulus-responsive soft material, thus avoiding the problems of large size and complex wiring caused by the physical integration of external sensors and driving units. At the same time, since the conductive coating is directly attached to the surface of the soft material, when the soft material deforms under external stimulation, the conductive coating can deform synchronously and output a continuous electrical signal without relying on any external sensing elements.

[0033] Specifically, refer to Figure 2 When the self-sensing artificial muscle is in a relaxed state, a certain gap is maintained between the conductive fiber layers, and the initial resistance measured at this time is R0. When the muscle contracts due to power, the liquid crystal elastomer fibers undergo thermal contraction, and the holes in the knot structure gradually close with the contraction, resulting in an increase in the contact area and the number of contact points between adjacent conductive fiber layers, thereby forming more parallel conductive paths. The overall resistance decreases from the initial value R0 to R0-ΔR (i.e., the resistance decreases). This sensing mechanism realizes the synchronous correspondence between the driven deformation and the resistance change, so that the real-time deformation state of the muscle can be inverted by monitoring the resistance value in real time, without the need for any external sensors.

[0034] In one feasible embodiment, the liquid crystal elastomer fiber can be prepared by: Liquid crystal elastomer fibers were prepared using template method or 3D printing method.

[0035] One method for preparing liquid crystal elastomer fibers using 3D printing is as follows: First, prepare the liquid crystal elastomer printing ink by weighing 100 parts by weight of nematic liquid crystal monomer (such as RM257), 0.8 parts by weight of photoinitiator (Irgacure 369), 50 parts by weight of tetrafunctional thiol ester (pentaerythritol tetra(3-mercaptopropionate, PETMP)) and 3 parts by weight of amine initiator (diethylamine, DPA). Add the above 100 parts by weight of materials to 250 parts by weight of dichloromethane solvent at room temperature, and stir on a magnetic stirrer at 500 rpm for 2-4 hours until all components are completely dissolved and mixed evenly to form a homogeneous, clear and transparent prepolymer solution.

[0036] The prepolymer solution was transferred to an open glass container and placed in a forced-air drying oven at 85°C for 48 hours. This allowed most of the dichloromethane solvent to evaporate slowly, while the amine initiator promoted a partial Michael addition reaction between the thiol and the liquid crystal monomer, resulting in a significant increase in solution viscosity and the formation of a liquid crystal elastomer prepolymer gel with a certain degree of viscoelasticity (i.e., printing ink).

[0037] The above-mentioned liquid crystal elastomer prepolymer gel was poured into a pneumatic extrusion 3D printing system equipped with a conical needle (300 μm inner diameter). The air pressure was adjusted to 80 psi, and the needle was controlled to print linearly on the substrate at a speed of 2 mm / s to obtain a fiber precursor with uniform diameter. Immediately after printing, the fiber precursor was placed under a 395 nm ultraviolet light source with an intensity of 15 mW / cm² for 40 minutes to cure. The ultraviolet light excited the photoinitiator to initiate a rapid and thorough photoclick chemical reaction between the thiol and the liquid crystal monomer, forming a fully cross-linked three-dimensional network structure, thus obtaining the cured liquid crystal elastomer fiber.

[0038] One method for preparing liquid crystal elastomer fibers using the template method is as follows: inject liquid crystal elastomer prepolymer into a silicone template with micropores (approximately 200-500 μm in diameter), degas under vacuum, heat cure (e.g., cure at 80°C for 4 hours), and then demold to obtain liquid crystal elastomer fibers with uniform diameter.

[0039] Furthermore, the soft materials coated with conductive coatings are assembled or arranged to form one or more fiber bundles, i.e., conductive fiber bundles. That is, a conductive fiber bundle is a bundle-like structure composed of one or more fibrous materials whose surfaces have been coated with conductive coatings.

[0040] In one feasible embodiment, the conductive fiber bundle includes single-strand conductive fiber bundles and multi-strand conductive fiber bundles, and the method of fabricating the soft material coated with a conductive coating into conductive fiber bundles can be as follows: A single strand of soft material coated with a conductive coating is used as a single strand of conductive fiber bundle; and / or, multiple strands of soft material coated with a conductive coating are combined into a multi-strand conductive fiber bundle by arranging them in parallel, twisting or braiding.

[0041] For example, a single liquid crystal elastomer fiber (approximately 300 μm in diameter and 15 cm in length) with a surface coated with a silver nanowire conductive coating can be directly used as a conductive fiber bundle to form a single conductive fiber bundle.

[0042] Alternatively, take 10 identical liquid crystal elastomer fibers coated with a conductive coating, align the 10 liquid crystal elastomer fibers in parallel and bring them close together to form a bundle of multiple conductive fibers arranged in parallel; or divide the above 10 liquid crystal elastomer fibers into three groups and braid them into a flat braid bundle; or apply a certain twist (such as 5-10 twists / cm) to the 10 fibers as a whole to form a tightly wrapped twisted fiber bundle.

[0043] By bundling single or multiple fibers coated with conductive coatings, the overall cross-sectional area and driving force output of artificial muscles can be increased (multiple fibers connected in parallel can superimpose contractile forces). It can also provide a suitable size and mechanical basis for subsequent weaving into a knotted structure with deformation gaps. At the same time, the conductive layers of multiple fibers participate in signal transmission, which helps to smooth the signal with resistance changes.

[0044] Conductive fiber bundles are woven according to a preset weaving method, and the tightness between the fiber bundles is controlled during the weaving process, so that the knot structure after weaving retains deformation gaps that can be deformed, such as the distance between the loop holes of the conductive fiber bundles and the gap between two conductive fiber bundles; thereby forming a self-sensing artificial muscle.

[0045] In one feasible implementation, the size of the deformation gap is determined based on the sensing sensitivity requirement.

[0046] Among them, the sensing sensitivity is the magnitude of the relative change rate of resistance (ΔR / R0) caused by a unit deformation. The sensing sensitivity reflects the ability of the self-sensing artificial muscle to convert mechanical deformation into electrical signals.

[0047] During the process of weaving conductive fiber bundles into a knot structure, the specific dimensions of the deformation gap reserved in the knot structure can be adjusted and determined according to the preset sensing sensitivity requirements. Specifically, if high sensitivity is required (i.e., a small deformation can cause a significant change in resistance), a smaller initial gap is selected; if a wider measurement range is required (i.e., the ability to detect a larger range of deformations), a larger initial gap is selected.

[0048] By designing and adjusting the size of the deformation gap according to the sensing sensitivity requirements, the sensing performance of the fabricated self-sensing artificial muscle can be designed and controlled. For applications requiring precise control (e.g., micromanipulation robots), a smaller gap can be selected to obtain a highly sensitive feedback signal; for applications requiring a large deformation range (e.g., soft actuators), a larger gap can be selected to ensure that the resistance change remains monotonic and detectable throughout the entire deformation range.

[0049] In one feasible embodiment, the knot structure is at least one of a square knot, a reverse knot, a figure-eight knot, a single knot, a lark's head knot, and a three-strand braid. Because the square knot is simple and easy to prepare, and its deformation gap (i.e., the distance between the fold point and the square knot) is easy to precisely control and repeatedly adjust, it can generate a regular, monotonous gap closure process during driven contraction, thereby obtaining a stable and calibrable resistance change signal. In contrast, the reverse knot, figure-eight knot, and other structures are relatively complex, with uneven gap distribution, and multi-strand braids may introduce excessive inter-fiber friction, affecting signal reproducibility. Therefore, in this embodiment, the preferred knot structure is a square knot.

[0050] Among them, the square knot is a rope knot formed by using a line or object as an axis and weaving the two ends of another line around the axis. The structural feature of the square knot is that the knot is flat and symmetrical, and it is formed by two half knots in opposite directions interlocking. The reverse hand knot is formed by wrapping one end of the rope around oneself, and then passing the rope end through the loop and tightening it. The figure-eight knot is a structure that forms an "8" shape after being tied. The single knot is formed by wrapping the rope end around oneself to form a loop, and then passing the rope end through the loop from below and tightening it. The lark's head knot is a loop knot composed of two reverse half loop knots. The lark's head knot is formed by folding the middle part of the rope to form a loop, passing the loop from below the object and looping it over the object, and then pulling the two free ends out of the loop and tightening it. The three-strand braid is a braided structure in which three strands of conductive fiber are interwoven in sequence to form a braid.

[0051] In one feasible embodiment, the conductive fiber bundle comprises a single strand of conductive fiber bundle. When the conductive fiber bundle is a single strand of conductive fiber bundle, the knot structure is a square knot. The method of weaving the conductive fiber bundle into a knot structure with deformation gaps can be: The single strand of conductive fiber is folded in half, and the two free ends formed after folding are tied into a flat knot, with a predetermined gap between the flat knot and the folding point, to form a knot structure with a deformation gap.

[0052] Specifically, a single liquid crystal elastomer fiber coated with a conductive coating can be used as a single conductive fiber bundle. The single conductive fiber bundle is folded in half, and then the two free ends formed after folding are tied into a flat knot. The flat knot and the folding point are kept at a distance of 2-10 mm, which results in a loose knot structure, thus completing the preparation of a single self-sensing artificial muscle.

[0053] If the spacing is too small (e.g., less than 2 mm), the fiber bundles will be in close contact in the initial state, and no significant resistance change will be generated during the shrinkage process, resulting in insufficient sensing sensitivity. If the spacing is too large (e.g., greater than 10 mm), a larger shrinkage strain is required to achieve gap closure, which leads to increased driving power consumption and a narrower linear range. In order to retain sufficient deformation space in the initial relaxation state, the spacing between the flat knot and the folding point is preferably 6 mm in this embodiment.

[0054] In this embodiment, after the conductive fiber bundles are woven into a knot structure with deformation gaps to form a self-sensing artificial muscle, the following can also be done: The self-sensing artificial muscle undergoes cyclic pre-stretching training, which includes repeatedly stretching and releasing the artificial muscle at a frequency of 0.1Hz-1Hz and a strain amplitude of 10%-40% until the maximum difference in the relative rate of change of resistance between two adjacent cycles is less than 5%. This stabilizes the contact interface between the conductive coatings and establishes a repeatable strain-resistance relationship. After pre-stretching training, the contact state between the conductive fiber bundles tends to be consistent, and the initial resistance drift and cyclic hysteresis are significantly reduced (experiments show fluctuation <5% after 100 cycles), allowing users to obtain a reliable strain-resistance relationship without self-calibration.

[0055] In this embodiment, the cyclic pre-stretching training may specifically be: The self-sensing artificial muscle is placed in a constant temperature environment of 25℃-35℃. 200 continuous stretching and releasing operations are performed on the self-sensing artificial muscle at a constant frequency of 0.2 Hz and a constant strain amplitude of 20%. During these 200 operations, the curves corresponding to the relative change rate of resistance and strain are recorded every 50 operations. Training is considered complete when the average deviation between the curves of the last 50 operations and the first 50 operations is less than 3%. If the average deviation is greater than or equal to 3%, an additional 100 operations are performed using a frequency sweep of 0.1 Hz-0.5 Hz and an amplitude variation of 15%-30%, until the deviation requirement is met.

[0056] Through constant temperature and amplitude cyclic pre-stretching training, signal dispersion caused by differences in initial contact state between different batches of muscles can be eliminated in batches; adaptive frequency sweep amplitude additional training can be used to strengthen the training of individual stubborn samples, ensuring that all products meet the high repeatability standard, thereby realizing the industrialized and consistent production of self-sensing artificial muscles.

[0057] This embodiment presents the following performance tests and application verifications of the self-sensing artificial muscle prepared using the above method: One end of the self-sensing artificial muscle prepared according to the above steps is fixed to the test bracket, and a 1 g weight is suspended from the other end to provide initial tension. The two lead wires of the muscle are connected to a digital source meter (such as Keithley2450) to apply driving current and measure the resistance value in real time.

[0058] Drive performance test: A step current of 0.4 A was applied to the self-sensing artificial muscle; the current flowed through the conductive coating and generated Joule heating, which caused the internal liquid crystal elastomer fibers to heat up and undergo liquid crystal phase transition. The muscle produced a contraction strain of about 45% within 5 seconds; after the current was removed, the muscle cooled naturally, the liquid crystal molecules returned to their orderly arrangement, and the muscle gradually returned to its original length; the above operation was repeated multiple times, and the drive response showed good reversibility.

[0059] Self-sensing performance verification: The initial resistance R0 of the muscle was measured to be 13Ω in the muscle relaxation state; when a driving current of 0.4 A was applied to cause muscle contraction, the resistance decreased steadily as the muscle length shortened, and the resistance dropped to 55% of the initial value (i.e., about 7.15Ω) when the maximum contraction strain was reached; the resistance change occurred synchronously with the length change, indicating that the self-sensing artificial muscle can reflect the deformation state in real time through its own resistance response without the need for an external sensor, showing excellent sensing responsiveness.

[0060] Repeatability and stability testing: Cyclic drive tests were conducted with a drive current of 0.4 A at different frequencies (0.1 Hz, 0.2 Hz, 0.4 Hz, and 0.675 Hz) and different strain amplitudes (10%, 20%, and 30%). The contraction strain and corresponding resistance change were recorded in each cycle. The experimental results show that after 100 consecutive cycles, the fluctuation value of the resistance change is less than 5%. This indicates that the self-sensing artificial muscle has excellent drive repeatability and sensing stability under long-term cyclic drive conditions, and can meet the requirements of reliable closed-loop control in applications such as soft robots and bionic devices.

[0061] In this embodiment, the driving function of liquid crystal elastomer fiber and the sensing function of surface conductive fiber layer are integrated into one. The deformation state of muscle can be synchronously obtained by real-time monitoring of resistance changes without the need for external strain gauges or vision systems, realizing closed-loop control of driving and sensing. The process is simple, highly repeatable, and suitable for large-scale production.

[0062] Based on the first embodiment of this application, a method for preparing a self-sensing artificial muscle is also provided. The same or similar content as the above embodiment can be referred to the above description, and will not be repeated hereafter. Specifically, the self-sensing artificial muscle is prepared using the method for preparing a self-sensing artificial muscle as described above.

[0063] In this embodiment, the self-sensing artificial muscle is used as a flexible actuator, an actuation component of a soft robot, or a power and sensing element in a wearable device.

[0064] For example, in flexible actuator applications, self-sensing artificial muscles can be used as driving elements and installed in devices such as flexible robotic arms, soft grippers, or biomimetic peristaltic robots. Specifically, the two ends of the artificial muscle are fixed to the fixed end and the movable end of the actuator, respectively. A driving current is applied through a control circuit to cause the muscle to contract, thereby driving the movable end to produce displacement or force output.

[0065] In soft robot applications, this self-sensing artificial muscle can be installed at the robot's joints. By applying a driving current, its contraction and relaxation can be controlled, thereby driving the robot to perform actions such as grasping, crawling, or bending. At the same time, the actual motion state of the joint can be obtained by monitoring the resistance change of the muscle in real time. There is no need to install additional angle or force sensors, which simplifies the control system structure and improves the reliability of operation in complex environments (such as underwater or confined spaces).

[0066] In wearable device applications, this self-sensing artificial muscle can be integrated into exoskeletons or rehabilitation gloves, providing both assistive drive and real-time monitoring of the user's limb movement intentions or muscle status via resistive signals, thus achieving closed-loop control for human-machine collaboration.

[0067] The above-described application methods in this embodiment all utilize the core characteristics of the integrated actuation and sensing of self-sensing artificial muscles, and have significant advantages such as compact structure, fast response, and accurate feedback.

[0068] Based on the first embodiment of this application, a method for controlling a self-sensing artificial muscle is also provided. Contents identical or similar to those in the above embodiment can be found in the above description and will not be repeated hereafter. Specifically, the self-sensing artificial muscle is prepared using the self-sensing artificial muscle preparation method described above.

[0069] Specifically, the control method for the self-sensing artificial muscle includes: In response to a drive command, an electrical signal is applied to both ends of the self-sensing artificial muscle, and the resistance values ​​at both ends of the self-sensing artificial muscle are collected simultaneously. Based on the resistance value and a preset relationship model between the change in muscle length and the relative rate of change in resistance of the artificial muscle, the current muscle length of the self-sensing artificial muscle is determined. Based on the current muscle length, the electrical signal is adjusted to control the deformation of the self-sensing artificial muscle.

[0070] In one feasible implementation, repeated excitation experiments can be conducted by applying currents of different amplitudes (e.g., 0.1A~0.6A), frequencies (e.g., 0.1Hz~1Hz), or waveforms (e.g., square waves, triangular waves) to the self-sensing artificial muscle. Simultaneously, the corresponding length change signals (measured with the aid of a laser displacement meter or visual markers) and resistance change signals are sampled and recorded. Based on the collected dataset, a stable and repeatable empirical correspondence model between the muscle length change ΔL and the relative rate of change of resistance ΔR / R0 is constructed through mathematical fitting (e.g., least squares method), i.e., ΔL = f(ΔR / R0). The relationship model is embedded in the memory of the control system in the form of a lookup table or polynomial coefficients and retrieved during real-time control.

[0071] It should be noted that the driving command can be a command signal issued by the control system (such as a microcontroller, computer, or dedicated drive circuit) to initiate or regulate the driving behavior of the artificial muscle. It typically includes information such as the target deformation, driving current amplitude, and duration. The electrical signal is the electrical energy input applied to both ends of the self-sensing artificial muscle. It can be direct current, pulsed current, voltage, or an electrical excitation with a specific waveform, used to generate Joule heating to drive muscle contraction. The resistance value is the DC resistance or AC impedance between the two ends of the self-sensing artificial muscle. The resistance value depends on the contact state between the conductive coatings and varies with muscle deformation.

[0072] Upon receiving the drive command, the control system applies a preset electrical signal (e.g., a step current with an amplitude of 0.4A) to both ends of the artificial muscle via the drive circuit. Simultaneously, it continuously acquires the voltage and current at both ends of the muscle using a high-precision analog-to-digital converter, calculating the resistance value or directly reading the resistance value, thus achieving synchronous drive and measurement. Since the resistance value can reflect the deformation state of the artificial muscle in real time, synchronous acquisition ensures that there is no time delay between the drive response and the sensor feedback, providing a high-time-resolution input signal for subsequent closed-loop control.

[0073] The relative rate of change of resistance ΔR / R0 is calculated based on the real-time measured resistance value R and the initial resistance R0. This relative rate of change is then substituted into the pre-calibrated relational model ΔL = f(ΔR / R0) to calculate the current change in muscle length ΔL, thus obtaining the current muscle length L = L0 + ΔL (L0 being the initial length). Without any external displacement sensors or vision systems, the real-time deformation state of the artificial muscle can be accurately determined solely through electrical measurements. This relational model, obtained through experimental calibration, exhibits high repeatability and low hysteresis (examples show resistance fluctuation <5% after 100 cycles), ensuring the reliability and accuracy of the length information obtained.

[0074] The current muscle length can be compared with the target length in the drive command to calculate the error signal. The control system can adjust the applied electrical signal according to the magnitude and direction of the error (e.g., increase the current to further contract, or decrease / remove the current to allow the muscle to relax), so that the muscle length approaches the target length, forming a closed-loop negative feedback control.

[0075] By dynamically adjusting the drive signal using real-time sensed deformation as feedback, precise closed-loop control of the artificial muscle is achieved without the need for external sensors. This not only simplifies the system structure and improves reliability in complex environments such as obstruction and underwater conditions, but also compensates for nonlinear behavior caused by load variations or material fatigue, ensuring long-term operational stability.

[0076] Reference Figure 3 This paper presents the curves showing the relative rate of change of electrical resistance (ΔR / R0) of a self-sensing artificial muscle over time under multiple cycles of actuation. Curves (a), (b), and (c) represent the test results at different numbers of cycles (e.g., the 1st, 50th, and 100th cycles), respectively. Figure 3 As can be seen, under the condition of applying the same driving current (0.4 A), the ΔR / R0 curves under different cycles basically overlap, and the maximum fluctuation is less than 5%.

[0077] Specifically, during the contraction phase, ΔR / R0 steadily decreased to approximately -0.45 (corresponding to a 45% resistance decrease rate), and during the recovery phase, ΔR / R0 smoothly rebounded to its initial value without significant drift or hysteresis. This result demonstrates that the self-sensing artificial muscle fabricated using the above method exhibits excellent repeatability and stability of its sensing signals during long-term cyclic actuation. The correlation between resistance changes and muscle deformation is highly reproducible, providing a reliable feedback basis for closed-loop control without external sensors.

[0078] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the preparation method of the self-sensing artificial muscle of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0079] The above descriptions are merely some embodiments of this application and do not limit the scope of protection of this application. Any equivalent structural transformations made based on the technical concept of this application and the content of this specification and drawings, or direct / indirect applications in other related technical fields, are included within the scope of protection of this application. All actions involving the acquisition of signals, information, or data in this application are performed in accordance with the relevant data protection laws and policies of the country where the application is located and with authorization from the owner of the corresponding device.

Claims

1. A method for preparing a self-sensing artificial muscle, characterized by, The method for preparing the self-sensing artificial muscle includes: A conductive coating is applied to the surface of a stimulus-responsive soft material; Soft materials coated with a conductive coating are made into conductive fiber bundles; The conductive fiber bundles are woven into a knotted structure with deformation gaps to form a self-sensing artificial muscle.

2. The method for preparing self-sensing artificial muscle as described in claim 1, characterized in that, The size of the deformation gap is determined based on the sensing sensitivity requirements.

3. The method for preparing self-sensing artificial muscle as described in claim 1, characterized in that, The knot structure is at least one of the following: square knot, reverse knot, figure-eight knot, single knot, lark's head knot, and three-strand braid.

4. The method for preparing self-sensing artificial muscle as described in claim 3, characterized in that, The conductive fiber bundle includes a single-strand conductive fiber bundle. When the conductive fiber bundle is a single-strand conductive fiber bundle, the knot structure is a square knot. The step of weaving the conductive fiber bundle into a knot structure with deformation gaps includes: The single strand of conductive fiber is folded in half, and the two free ends formed after folding are tied into a flat knot, with a predetermined gap between the flat knot and the folding point, to form a knot structure with a deformation gap.

5. The method for preparing self-sensing artificial muscle as described in claim 1, characterized in that, The conductive fiber bundle includes single-strand conductive fiber bundles and multi-strand conductive fiber bundles. The step of fabricating the soft material coated with a conductive coating into a conductive fiber bundle includes: A single strand of soft material coated with a conductive coating is used as a single-strand conductive fiber bundle. And / or, multiple strands of soft material coated with conductive coating are combined into a multi-strand conductive fiber bundle by arranging them in parallel, twisting or braiding.

6. The method for preparing self-sensing artificial muscle as described in claim 1, characterized in that, The stimulus-responsive soft material is either liquid crystal elastomer fiber or hydrogel smart material fiber.

7. The method for preparing self-sensing artificial muscle as described in claim 1, characterized in that, The conductive coating material includes at least one of stainless steel soft material, conductive polymer, carbon nanomaterial and metal nanowire.

8. A method for applying a self-sensing artificial muscle, wherein the self-sensing artificial muscle is prepared using the preparation method of the self-sensing artificial muscle as described in any one of claims 1 to 7, characterized in that, The self-sensing artificial muscle can be used as a flexible actuator, an actuation component of a soft robot, or a power and sensing element in a wearable device.

9. A method for controlling a self-sensing artificial muscle, wherein the self-sensing artificial muscle is prepared using the method for preparing a self-sensing artificial muscle as described in any one of claims 1 to 7, characterized in that, The control method for the self-sensing artificial muscle includes: In response to the driving command, electrical signals are applied to both ends of the self-sensing artificial muscle, and the resistance values ​​at both ends of the self-sensing artificial muscle are collected synchronously. Based on the resistance value and a preset relationship model between the change in muscle length of the artificial muscle and the relative rate of change in resistance, the current muscle length of the self-sensing artificial muscle is determined. The electrical signal is adjusted based on the current muscle length to control the deformation of the self-sensing artificial muscle.

Images