Dielectric elastomer sensor and actuator integrated system for soft robots

Abstract

The present application relates to the technical field of soft robot, and particularly relates to a dielectric elastomer sensing and driving integrated system for a soft robot, which comprises a dielectric elastomer array module, a periodic signal acquisition module, a timing scheduling processing module and a driving voltage output module. The dielectric elastomer array module is used for switching between a sensing mode and a driving mode under the control of the timing scheduling processing module. The periodic signal acquisition module is used for collecting an environmental periodic disturbance signal and predicting a next disturbance peak arrival time. The timing scheduling processing module is used for switching the dielectric elastomer unit to the sensing mode in advance according to the predicted time, collecting capacitance data and solving contact state data, and switching a closed-loop force control mode and a low stiffness servo mode correspondingly. The driving voltage output module is used for generating driving voltage by fusing a feedforward compensation component and a feedback correction component, performing breakdown margin judgment before output, and outputting a diagnosis data packet after each cycle. The present application realizes sensing and driving integration, and has the characteristics of strong anti-interference and high adaptive capacity.

Description

Technical Field

[0001] This invention relates to the field of soft robot technology, and more specifically to an integrated dielectric elastomer sensing and actuation system for soft robots. Background Technology

[0002] Soft robots, with their inherent compliance, environmental adaptability, and human-robot interaction safety, have shown broad application prospects in fields such as medical rehabilitation, bionic grasping, and complex environment exploration. Dielectric elastomers, as typical electroactive polymers, can undergo large deformations under electric field excitation, and these deformations can be converted into detectable capacitance changes, thus being regarded as ideal "artificial muscles" and sensing elements for soft robots. However, existing soft robot systems typically design sensing and actuation functions separately, relying on external sensors to obtain state information. This not only increases the complexity of the system structure and control but also limits the robot's response speed and integration in dynamic, unstructured environments.

[0003] Currently, attempts to use dielectric elastomers for both sensing and actuation often employ time-division multiplexing strategies. However, several key issues arise in practical applications: First, the external environment frequently experiences periodic disturbances. If the sensing sampling time does not match the peak phase of the disturbance, it becomes difficult to accurately capture the contact state, leading to control lag. Second, dielectric elastomers are susceptible to temperature drift and baseline capacitance changes during long-term operation, reducing sensing accuracy. Third, during interaction with the outside world, the interface contact state of soft robots may change abruptly, and traditional fixed control modes cannot be flexibly switched, easily causing control instability or damage to the contacted object. These shortcomings severely restrict the reliability and intelligence level of soft robots driven by dielectric elastomers.

[0004] To address the aforementioned issues, this invention proposes an integrated dielectric elastomer sensing and actuation system for soft robots, which achieves highly integrated, robust, and adaptive sensing and actuation control, effectively improving the operational performance of soft robots in complex dynamic environments. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides an integrated dielectric elastomer sensing and driving system for soft robots, which can effectively solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] This invention provides an integrated dielectric elastomer sensing and actuation system for soft robots, comprising:

[0008] The dielectric elastomer array module includes multiple independent dielectric elastomer units, each of which alternates between sensing mode and driving mode under the control of the timing scheduling processing module.

[0009] The periodic signal acquisition module is used to acquire characteristic signals that characterize the periodic disturbances in the environment in which the system is located, calculate the predicted arrival time of the next disturbance peak based on the characteristic signals, and send the predicted arrival time to the timing scheduling processing module.

[0010] The timing scheduling processing module is used to switch the dielectric elastomer units in driving mode to sensing mode before the predicted arrival time, according to the predicted arrival time, to acquire the capacitance-related data of each dielectric elastomer unit, and to calculate the contact state data of each dielectric elastomer unit based on the capacitance-related data; the timing scheduling processing module identifies whether a sudden change in interface state has occurred based on the changes in the contact state data, and switches the working mode of the corresponding dielectric elastomer unit when a sudden change in interface state is detected; the working modes include closed-loop force control mode and low stiffness follow-up mode.

[0011] The driving voltage output module is used to generate a feedforward compensation component based on the predicted arrival time, generate a feedback correction component based on the deviation between the contact state data and the preset target contact state data, fuse the feedforward compensation component and the feedback correction component to obtain the driving voltage of each dielectric elastomer unit and output it; the driving voltage output module determines whether the driving voltage exceeds the preset breakdown margin voltage before outputting the driving voltage, and if it exceeds the preset breakdown margin voltage, it cuts off the output voltage to the preset breakdown margin voltage.

[0012] Furthermore, when the periodic signal acquisition module calculates the predicted arrival time of the next disturbance peak based on the characteristic signal, it is specifically used for:

[0013] The timestamps of adjacent characteristic peaks in the characteristic signal are extracted, and the difference between adjacent timestamps is calculated to obtain the perturbation period length. The average of the perturbation period lengths for a preset window number is calculated to obtain the period average. Combining the preset phase lead coefficient and the timestamp of the most recent characteristic peak, the predicted arrival time satisfies the following calculation relationship:

[0014] ;

[0015] in, To calculate the predicted arrival time of the next disturbance peak; The most recent characteristic peak timestamp; The phase lead coefficient preset for the system; The average length of the disturbance period within the preset window is used; the periodic signal acquisition module sends the calculated predicted arrival time to the timing scheduling processing module, which uses the predicted arrival time as a reference and sets the trigger time for the sensing mode switching at a preset advance time before the predicted arrival time.

[0016] Furthermore, when acquiring the capacitance-related data of each dielectric elastomer unit, the timing scheduling processing module is specifically used for:

[0017] In sensing mode, a detection voltage lower than a preset sensing voltage threshold is applied to each dielectric elastomer unit, and the real-time capacitance value of each dielectric elastomer unit is collected. The timing scheduling processing module determines whether the corresponding dielectric elastomer unit is continuously in a state without applied driving voltage within a preset silent window. If so, the dynamic baseline capacitance is updated with the average capacitance value within the preset silent window. If not, the dynamic baseline capacitance recorded when the condition was met is kept unchanged. The dynamic baseline capacitance is used to correct the real-time capacitance value in subsequent steps.

[0018] Furthermore, when the timing scheduling processing module calculates the contact state data based on the real-time capacitance value and the dynamic baseline capacitance, it specifically performs the following: performs a differential calculation between the real-time capacitance value and the corresponding dynamic baseline capacitance to obtain the net deformation capacitance value, wherein the net deformation capacitance value satisfies the following calculation relationship:

[0019] ;

[0020] in, To calculate the net deformation capacitance value of the dielectric elastomer unit in the i-th row and j-th column at time t; This represents the real-time capacitance value of the dielectric elastomer unit collected at time t. The dynamic baseline capacitance of the dielectric elastomer unit is used as an index input to query the pre-stored calibration lookup table, obtain the corresponding contact force value as the contact state data, and use the contact state data for subsequent interface state change identification and feedback correction component generation.

[0021] Furthermore, the pre-stored calibration lookup table is established by applying gradient pressure loads to each dielectric elastomer unit during the factory calibration stage, synchronously recording the correspondence between the net deformation capacitance value and the pressure value, and storing it in the system's non-volatile memory in groups according to temperature gradients; the timing scheduling processing module selects the calibration lookup table corresponding to the temperature group based on the temperature sensor readings during operation to eliminate the influence of working temperature changes on the accuracy of contact force value mapping.

[0022] Furthermore, when identifying whether a sudden change in interface state has occurred, the timing scheduling processing module is specifically used for:

[0023] The rate of change of the contact state data of each dielectric elastomer unit relative to the previous sampling time is calculated. When the absolute value of the rate of change exceeds the preset mutation threshold, it is determined that the interface state mutation of the corresponding dielectric elastomer unit has occurred, and the dielectric elastomer unit is switched from closed-loop force control mode to low stiffness follow-up mode. When the absolute value of the rate of change of the dielectric elastomer unit falls back to below the preset stability threshold, it is determined that the interface state mutation has ended, and the dielectric elastomer unit is restored to closed-loop force control mode.

[0024] Furthermore, the timing scheduling processing module also calculates the spatial distribution dispersion coefficient for the contact state data of all dielectric elastomer units in the closed-loop force control mode. The spatial distribution dispersion coefficient is the ratio of the standard deviation to the mean of the contact state data of the dielectric elastomer units. When the spatial distribution dispersion coefficient exceeds a preset dispersion threshold, the driving voltage output module increases the driving voltage of the dielectric elastomer units whose contact state data is lower than a preset minimum contact threshold to a preset compensation voltage value. Dielectric elastomer units in the low stiffness follow-up mode do not participate in the calculation of the spatial distribution dispersion coefficient.

[0025] Furthermore, when generating the driving voltage based on the feedforward compensation component and the feedback correction component, the driving voltage output module is specifically used to: superimpose the feedforward compensation component and the feedback correction component to obtain the driving voltage of each dielectric elastomer unit, wherein the driving voltage satisfies the following calculation relationship:

[0026] ;

[0027] in, To calculate the driving voltage of the dielectric elastomer unit in the i-th row and j-th column at time t; For the feedforward compensation component of the dielectric elastomer unit; This is the feedback correction component for the dielectric elastomer unit; the driving voltage output module will calculate the... As the driving voltage output.

[0028] Furthermore, the feedforward compensation component is determined based on the weighted average of the displacement amplitudes of the corresponding dielectric elastomer units in the historical disturbance cycles within a preset lead time period. The weights decrease with time towards the historical direction according to a preset attenuation coefficient, so that the displacement amplitudes corresponding to the disturbance cycles closer to the current time are given higher weights. The feedforward compensation component is continuously updated based on the weighted average as the system operates to adapt to the drift of disturbance amplitudes caused by changes in external operating conditions.

[0029] Furthermore, after completing each sensing sampling and drive compensation cycle, the drive voltage output module outputs a periodic diagnostic data packet to the host computer. The periodic diagnostic data packet includes the contact state data of each dielectric elastomer unit in this cycle, the difference between the predicted arrival time and the actual disturbance peak arrival time, and the number of the dielectric elastomer unit that switched operating modes. The host computer corrects the preset phase lead coefficient used by the periodic signal acquisition module based on the difference, so that the accuracy of the predicted arrival time gradually converges as the number of system runs increases.

[0030] The technical solution provided by this invention has the following advantages compared with the known prior art:

[0031] This invention predicts the peak value of periodic environmental disturbances and switches the dielectric elastomer unit from driving mode to sensing mode before the disturbance arrives, ensuring precise phase alignment between the sensing sample and the disturbance. This effectively overcomes the control lag problem caused by phase mismatch in existing technologies. Simultaneously, the system generates a feedforward compensation component based on the predicted time, enabling the soft robot to respond proactively before the disturbance arrives, thus improving its ability to suppress periodic disturbances. Furthermore, through a dynamic baseline capacitance update mechanism within a silent window and a calibration lookup table grouped by temperature, the influence of temperature drift and material relaxation on sensing accuracy is effectively eliminated, significantly improving the long-term stability of contact force sensing.

[0032] This invention monitors the rate of change of contact state data in real time. When sudden changes occur at the interface, such as slippage or collision, the dielectric elastomer unit adaptively switches from closed-loop force control mode to low-stiffness servo mode. This avoids the problem of damage to the contacted object or its own structure due to excessive stiffness under sudden changes, which is common in traditional fixed control modes. After the sudden change, closed-loop force control is automatically restored, balancing control accuracy and system safety. Simultaneously, the spatial distribution dispersion coefficient of the dielectric elastomer unit in closed-loop force control mode is calculated. When the contact force distribution is uneven, the driving voltage in low contact force areas is automatically increased, making the overall contact force of the array more uniform and improving the grasping stability and operational reliability of the soft robot. The system also achieves adaptive correction of the phase lead coefficient through periodic diagnostic data, forming a self-learning closed loop, further improving environmental adaptability and prediction accuracy during long-term operation. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0034] Figure 1This is a schematic diagram of the system structure of the present invention;

[0035] Figure 2 This is a schematic diagram of the data flow in this invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0037] The present invention will be further described below with reference to embodiments.

[0038] Example:

[0039] Reference Figure 1 , Figure 2 This embodiment provides an integrated dielectric elastomer sensing and driving system for soft robots. The system is integrated and deployed on the soft robot body to simultaneously realize driving control and contact state perception during the periodic or non-periodic contact between the soft robot and the external environment. Specifically, it includes a dielectric elastomer array module, a periodic signal acquisition module, a timing scheduling and processing module, and a driving voltage output module.

[0040] The dielectric elastomer array module includes multiple independent dielectric elastomer units, each of which alternates between sensing mode and driving mode under the control of the timing scheduling processing module.

[0041] In this embodiment, each dielectric elastomer unit is composed of a flexible electrode layer and a dielectric elastomer film stack. Multiple dielectric elastomer units are arranged in rows and columns to form an array, distributed on the surface or inside the soft robot body in areas prone to deformation and contact. At any given time, each dielectric elastomer unit is only in one of two modes: sensing mode and driving mode. In driving mode, the dielectric elastomer unit deforms under the action of driving voltage to realize the movement or grasping action of the soft robot. In sensing mode, the dielectric elastomer unit is used to acquire its own capacitance-related data. The specific electrode lead-out method and packaging structure of the dielectric elastomer array module can be implemented using conventional flexible circuit printing and stacked packaging processes in the art, and this invention does not limit this.

[0042] The periodic signal acquisition module is used to acquire characteristic signals that characterize the periodic disturbances in the environment in which the system is located, calculate the predicted arrival time of the next disturbance peak based on the characteristic signals, and send the predicted arrival time to the timing scheduling processing module.

[0043] In this embodiment, the feature signal can be continuously collected by an auxiliary sensing unit (such as an accelerometer or a dielectric elastomer unit specifically designed for monitoring environmental disturbances) independently set on the soft robot body, in order to reflect the fluctuation pattern of periodic disturbances in the environment in which the soft robot is located over time, such as periodic ground-touching disturbances during gait walking, periodic vibration disturbances during object grasping, etc.

[0044] In this embodiment, when the periodic signal acquisition module calculates the predicted arrival time of the next disturbance peak based on the characteristic signal, it is specifically used for:

[0045] Extract the timestamps of adjacent characteristic peaks from the feature signal, calculate the difference between adjacent timestamps to obtain the perturbation period length, and calculate the average of the perturbation period lengths for a preset number of windows to obtain the period mean. Combine the preset phase lead coefficient with the timestamp of the most recent characteristic peak, and predict the arrival time according to the following calculation formula:

[0046] ;

[0047] in, To calculate the predicted arrival time of the next disturbance peak; The most recent characteristic peak timestamp; The phase lead coefficient preset for the system; The average length of the disturbance period within the preset window is used; the periodic signal acquisition module sends the calculated predicted arrival time to the timing scheduling processing module, and sets the trigger time for the sensor mode switching to a preset advance time before the predicted arrival time, based on the predicted arrival time.

[0048] The periodic signal acquisition module sends the calculated predicted arrival time to the timing scheduling processing module. The timing scheduling processing module then sets the trigger time for sensor mode switching to a preset lead time prior to the predicted arrival time, using the predicted arrival time as a reference. Through this prediction method based on historical periodic averages and phase lead coefficients, mode switching can be completed before external periodic disturbances arrive, avoiding the problem of delayed contact state capture due to phase mismatch between the sensor sampling time and the disturbance peak.

[0049] The timing scheduling processing module is used to switch the dielectric elastomer units in driving mode to sensing mode before the predicted arrival time, based on the predicted arrival time. It acquires the capacitance-related data of each dielectric elastomer unit and calculates the contact state data of each dielectric elastomer unit based on the capacitance-related data. The timing scheduling processing module identifies whether a sudden change in interface state has occurred based on the changes in the contact state data, and switches the working mode of the corresponding dielectric elastomer unit when a sudden change in interface state is detected. The working modes include closed-loop force control mode and low stiffness follow-up mode.

[0050] In this embodiment, the timing scheduling processing module can be implemented by a microprocessor or digital signal processing chip with real-time computing capabilities. It is connected to the periodic signal acquisition module, the dielectric elastomer array module and the driving voltage output module through a signal bus, and is used to uniformly schedule the working mode switching, data calculation and control strategy output of each dielectric elastomer unit.

[0051] In this embodiment, when the timing scheduling processing module acquires the capacitance-related data of each dielectric elastomer unit, it specifically performs the following:

[0052] In sensing mode, a detection voltage lower than the preset sensing voltage threshold is applied to each dielectric elastomer unit, and the real-time capacitance value of each dielectric elastomer unit is collected. The timing scheduling processing module determines whether the corresponding dielectric elastomer unit is continuously in a state without driving voltage application within a preset silent window. If so, the dynamic baseline capacitance is updated with the average capacitance value within the preset silent window. If not, the dynamic baseline capacitance recorded when the condition was met is kept unchanged. The dynamic baseline capacitance is used to correct the real-time capacitance value in subsequent steps.

[0053] The specific application method of the detection voltage and the capacitance acquisition circuit can be implemented using conventional capacitance detection circuits in the field (such as capacitance detection circuits based on the charge-discharge method or the bridge method), and this invention does not limit this. The above-mentioned dynamic baseline capacitance update mechanism enables the baseline capacitance to be adaptively updated according to factors such as creep and temperature drift that occur during long-term operation of the dielectric elastomer material, thereby avoiding systematic deviations in subsequent contact force calculations due to baseline drift.

[0054] In this embodiment, when the timing scheduling processing module calculates the contact state data based on the real-time capacitance value and the dynamic baseline capacitance, it specifically performs the following: It performs a differential calculation between the real-time capacitance value and the corresponding dynamic baseline capacitance to obtain the net deformation capacitance value, which satisfies the following calculation relationship:

[0055] ;

[0056] in, To calculate the net deformation capacitance value of the dielectric elastomer unit in the i-th row and j-th column at time t; This represents the real-time capacitance value of the dielectric elastomer unit collected at time t. The dynamic baseline capacitance of the dielectric elastomer unit is used as an index input. The timing scheduling processing module uses the net deformation capacitance value as an index input to query the pre-stored calibration lookup table, obtains the corresponding contact force value as contact state data, and uses the contact state data for subsequent interface state change identification and feedback correction component generation.

[0057] The timing scheduling processing module uses the net deformation capacitance value as an index input to query the pre-stored calibration lookup table, obtains the corresponding contact force value as contact state data, and uses the contact state data for subsequent interface state change identification and feedback correction component generation.

[0058] In this embodiment, the pre-stored calibration lookup table is established during the factory calibration stage by applying gradient pressure loads to each dielectric elastomer unit and synchronously recording the correspondence between the net deformation capacitance value and the pressure value. This table is then grouped by temperature gradient and stored in the system's non-volatile memory. During runtime, the timing scheduling module selects the calibration lookup table corresponding to the temperature group based on the temperature sensor readings to eliminate the influence of operating temperature changes on the accuracy of the contact force value mapping. The temperature sensor can be placed near the dielectric elastomer array module to obtain the ambient temperature of the array in real time. The specific storage structure and lookup method of the calibration lookup table can be implemented using conventional data table indexing methods in the art.

[0059] In this embodiment, when the timing scheduling processing module identifies whether a sudden change in interface state has occurred, it is specifically used for:

[0060] The rate of change of the contact state data of each dielectric elastomer unit relative to the previous sampling time is calculated. When the absolute value of the rate of change exceeds the preset mutation threshold, it is determined that the interface state mutation of the corresponding dielectric elastomer unit has occurred, and the dielectric elastomer unit is switched from closed-loop force control mode to low stiffness follow-up mode. When the absolute value of the rate of change of the dielectric elastomer unit falls back to below the preset stability threshold, it is determined that the interface state mutation has ended, and the dielectric elastomer unit is restored to closed-loop force control mode.

[0061] Among them, the dielectric elastomer unit in closed-loop force control mode performs closed-loop drive control according to the preset target contact state data to maintain a stable contact force; the dielectric elastomer unit in low stiffness follow-up mode reduces the adjustment amplitude of the corresponding drive voltage, making the mechanical response of the dielectric elastomer unit more compliant, thereby passively following the instantaneous deformation of the external contact object, avoiding excessive impact or damage to the contact object due to control response lag at the moment when the interface state changes abruptly (such as the contact object suddenly slipping, colliding or detaching).

[0062] In this embodiment, the timing scheduling processing module also calculates the spatial distribution dispersion coefficient for the contact state data of all dielectric elastomer units in the closed-loop force control mode. The spatial distribution dispersion coefficient is the ratio of the standard deviation to the mean of the contact state data of the dielectric elastomer units. When the spatial distribution dispersion coefficient exceeds the preset dispersion threshold, the driving voltage output module increases the driving voltage of the dielectric elastomer units whose contact state data is lower than the preset minimum contact threshold to the preset compensation voltage value. The dielectric elastomer units in the low stiffness follow-up mode do not participate in the calculation of the spatial distribution dispersion coefficient.

[0063] By calculating the spatial distribution dispersion coefficient and using the driving voltage compensation mechanism, the driving voltage of the dielectric elastomer unit corresponding to the insufficient contact area can be increased in a timely manner when the soft robot interacts with an irregularly shaped contact object and the forces on each dielectric elastomer unit are uneven. This makes the overall contact state more uniform and improves the stability of the soft robot's gripping or bonding operations.

[0064] The drive voltage output module is used to generate a feedforward compensation component based on the predicted arrival time, and to generate a feedback correction component based on the deviation between the contact state data and the preset target contact state data. The feedforward compensation component and the feedback correction component are fused to obtain the drive voltage of each dielectric elastomer unit and output it. Before outputting the drive voltage, the drive voltage output module determines whether it exceeds the preset breakdown margin voltage. If it does, it cuts off the output voltage to the preset breakdown margin voltage.

[0065] In this embodiment, the specific calculation method of the feedback correction component can adopt the conventional proportional-integral-derivative control method in the art, and is calculated based on the deviation between the contact state data and the preset target contact state data. This invention does not limit this method. The preset breakdown margin voltage is preset based on the breakdown field strength and film thickness of the dielectric elastomer material. The truncation process is used to prevent the driving voltage from exceeding the safe working voltage range of the material due to the superposition of the feedforward compensation component and the feedback correction component, which would cause electrical breakdown damage to the dielectric elastomer unit, thereby ensuring the reliability and safety of the system.

[0066] In this embodiment, when the driving voltage output module generates the driving voltage based on the feedforward compensation component and the feedback correction component, it is specifically used to: superimpose the feedforward compensation component and the feedback correction component to obtain the driving voltage of each dielectric elastomer unit. The driving voltage satisfies the following calculation relationship:

[0067] ;

[0068] in, To calculate the driving voltage of the dielectric elastomer unit in the i-th row and j-th column at time t; For the feedforward compensation component of the dielectric elastomer unit; The feedback correction component for the dielectric elastomer unit; the drive voltage output module will calculate the... As the driving voltage output.

[0069] The driving voltage output module uses the calculated driving voltage as the driving voltage output of the corresponding dielectric elastomer unit, and then applies it to the corresponding dielectric elastomer unit after the aforementioned breakdown margin judgment and cutoff processing.

[0070] In this embodiment, the feedforward compensation component is determined based on the weighted average of the displacement amplitudes of the corresponding dielectric elastomer units in the historical disturbance cycles within a preset lead time. The weights decrease with time toward the historical direction according to a preset attenuation coefficient, so that the displacement amplitudes corresponding to the disturbance cycles closer to the current time are given higher weights. The feedforward compensation component is continuously updated based on the weighted average as the system operates, in order to adapt to the drift of the disturbance amplitude caused by changes in external operating conditions.

[0071] By using the aforementioned feedforward compensation component generation method based on the weighted average of historical disturbance amplitudes, the drive voltage output module can apply a compensation drive voltage that matches the expected disturbance amplitude before the external disturbance arrives. Combined with the feedback correction component for closed-loop correction of the actual deviation, the drive voltage has both anticipatory response capability and closed-loop correction capability, thereby improving the response speed and control accuracy of the soft robot in a periodic disturbance environment.

[0072] In this embodiment, after completing each sensing sampling and drive compensation cycle, the drive voltage output module outputs a periodic diagnostic data packet to the host computer. The periodic diagnostic data packet includes the contact state data of each dielectric elastomer unit in this cycle, the difference between the predicted arrival time and the actual peak arrival time of the disturbance, and the number of the dielectric elastomer unit that has switched working modes. The host computer corrects the preset phase lead coefficient used by the periodic signal acquisition module based on the difference, so that the accuracy of the predicted arrival time gradually converges as the number of system runs increases.

[0073] The host computer can be an upper-level controller or computing unit that communicates with this system. The specific method by which it iteratively corrects the preset phase lead coefficient based on the received periodic diagnostic data packets can be implemented using conventional parameter self-tuning methods in this field, and this invention does not limit this approach. Through the output of the aforementioned periodic diagnostic data packets and the continuous correction of the phase lead coefficient, this system forms a closed-loop adaptive adjustment mechanism from prediction and execution to feedback correction, enabling the prediction accuracy of the periodic signal acquisition module for the arrival time of disturbance peaks to continuously improve with the accumulation of system operating time.

[0074] It should be noted that the switching between sensing and driving modes, the calculation of contact state data, the identification of interface state changes and the switching of working modes, and the generation of feedforward feedback of driving voltage for each of the above dielectric elastomer units are all achieved collaboratively by the timing scheduling processing module and the driving voltage output module in accordance with the aforementioned steps. This allows the system to simultaneously complete driving and sensing functions on a single dielectric elastomer array without the need for additional independent external sensors, effectively improving the integration of the soft robot system, the response speed in dynamic environments, and the safety and stability during contact interaction.

[0075] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. An integrated dielectric elastomer sensing and actuation system for soft robots, characterized in that, include: The dielectric elastomer array module includes multiple independent dielectric elastomer units, each of which alternates between sensing mode and driving mode under the control of the timing scheduling processing module. The periodic signal acquisition module is used to acquire characteristic signals that characterize the periodic disturbances in the environment in which the system is located, calculate the predicted arrival time of the next disturbance peak based on the characteristic signals, and send the predicted arrival time to the timing scheduling processing module. The timing scheduling processing module is used to switch the dielectric elastomer units in driving mode to sensing mode before the predicted arrival time, based on the predicted arrival time, to acquire capacitance-related data of each dielectric elastomer unit, and to calculate contact state data of each dielectric elastomer unit based on the capacitance-related data; the timing scheduling processing module identifies whether an interface state change has occurred based on the changes in the contact state data, and switches the operating mode of the corresponding dielectric elastomer unit when an interface state change is detected; the operating modes include closed-loop force control mode and low stiffness follow-up mode; when identifying whether an interface state change has occurred, the timing scheduling processing module is specifically used for: The rate of change of the contact state data of each dielectric elastomer unit relative to the previous sampling time is calculated. When the absolute value of the rate of change exceeds the preset mutation threshold, it is determined that the interface state mutation of the corresponding dielectric elastomer unit has occurred, and the dielectric elastomer unit is switched from closed-loop force control mode to low stiffness follow-up mode. When the absolute value of the rate of change of the dielectric elastomer unit falls back to below the preset stability threshold, it is determined that the interface state mutation has ended, and the dielectric elastomer unit is restored to closed-loop force control mode. The driving voltage output module is used to generate a feedforward compensation component based on the predicted arrival time, generate a feedback correction component based on the deviation between the contact state data and the preset target contact state data, fuse the feedforward compensation component and the feedback correction component to obtain the driving voltage of each dielectric elastomer unit and output it; the driving voltage output module determines whether the driving voltage exceeds the preset breakdown margin voltage before outputting the driving voltage, and if it exceeds the preset breakdown margin voltage, it cuts off the output voltage to the preset breakdown margin voltage.

2. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, When the periodic signal acquisition module calculates the predicted arrival time of the next disturbance peak based on the characteristic signal, it is specifically used for: The timestamps of adjacent characteristic peaks in the characteristic signal are extracted, and the difference between adjacent timestamps is calculated to obtain the perturbation period length. The average of the perturbation period lengths for a preset window number is calculated to obtain the period average. Combining the preset phase lead coefficient and the timestamp of the most recent characteristic peak, the predicted arrival time satisfies the following calculation relationship: ； in, To calculate the predicted arrival time of the next disturbance peak; The most recent characteristic peak timestamp; The phase lead coefficient preset for the system; The average length of the disturbance period within the preset window number is used; the periodic signal acquisition module sends the calculated predicted arrival time to the timing scheduling processing module, which uses the predicted arrival time as a reference and sets the trigger time for the sensing mode switching at a preset advance time before the predicted arrival time.

3. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, When the timing scheduling processing module acquires the capacitance-related data of each dielectric elastomer unit, it is specifically used for: In sensing mode, a detection voltage lower than a preset sensing voltage threshold is applied to each dielectric elastomer unit, and the real-time capacitance value of each dielectric elastomer unit is collected. The timing scheduling processing module determines whether the corresponding dielectric elastomer unit is continuously in a state without applied driving voltage within a preset silent window. If so, the dynamic baseline capacitance is updated with the average capacitance value within the preset silent window. If not, the dynamic baseline capacitance recorded when the condition was met is kept unchanged. The dynamic baseline capacitance is used to correct the real-time capacitance value in subsequent steps.

4. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 3, characterized in that, When the timing scheduling processing module calculates the contact state data based on the real-time capacitance value and the dynamic baseline capacitance, it specifically performs the following: performs a differential calculation between the real-time capacitance value and the corresponding dynamic baseline capacitance to obtain the net deformation capacitance value, wherein the net deformation capacitance value satisfies the following calculation relationship: ； in, To calculate the net deformation capacitance value of the dielectric elastomer unit in the i-th row and j-th column at time t; This represents the real-time capacitance value of the dielectric elastomer unit collected at time t. The dynamic baseline capacitance of the dielectric elastomer unit is used as an index input to query the pre-stored calibration lookup table, obtain the corresponding contact force value as the contact state data, and use the contact state data for subsequent interface state change identification and feedback correction component generation.

5. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 4, characterized in that, The pre-stored calibration lookup table is established by applying gradient pressure loads to each dielectric elastomer unit during the factory calibration stage, synchronously recording the correspondence between the net deformation capacitance value and the pressure value, and storing it in the system's non-volatile memory in groups according to temperature gradients; the timing scheduling processing module selects the calibration lookup table corresponding to the temperature group based on the temperature sensor readings during operation to eliminate the influence of working temperature changes on the accuracy of contact force value mapping.

6. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, The timing scheduling processing module also calculates the spatial distribution dispersion coefficient for the contact state data of all dielectric elastomer units in closed-loop force control mode. The spatial distribution dispersion coefficient is the ratio of the standard deviation to the mean of the contact state data of the dielectric elastomer units. When the spatial distribution dispersion coefficient exceeds a preset dispersion threshold, the driving voltage output module increases the driving voltage of the dielectric elastomer units whose contact state data is lower than a preset minimum contact threshold to a preset compensation voltage value. Dielectric elastomer units in low stiffness follow-up mode do not participate in the calculation of the spatial distribution dispersion coefficient.

7. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, When generating the driving voltage based on the feedforward compensation component and the feedback correction component, the driving voltage output module is specifically used to: superimpose the feedforward compensation component and the feedback correction component to obtain the driving voltage of each dielectric elastomer unit, wherein the driving voltage satisfies the following calculation relationship: ； in, To calculate the driving voltage of the dielectric elastomer unit in the i-th row and j-th column at time t; For the feedforward compensation component of the dielectric elastomer unit; This is the feedback correction component for the dielectric elastomer unit; the driving voltage output module will calculate the... As the driving voltage output.

8. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, The feedforward compensation component is determined based on the weighted average of the displacement amplitudes of the corresponding dielectric elastomer units in the historical disturbance cycles within a preset lead time. The weights decrease with time toward the historical direction according to a preset attenuation coefficient, so that the displacement amplitudes corresponding to the disturbance cycles closer to the current time are given higher weights. The feedforward compensation component is continuously updated based on the weighted average as the system operates, so as to adapt to the drift of disturbance amplitudes caused by changes in external operating conditions.

9. The integrated dielectric elastomer sensing and actuation system for soft robots according to claim 1, characterized in that, After completing each sensing sampling and drive compensation cycle, the drive voltage output module outputs a periodic diagnostic data packet to the host computer. The periodic diagnostic data packet includes the contact state data of each dielectric elastomer unit in this cycle, the difference between the predicted arrival time and the actual peak arrival time of the disturbance, and the number of the dielectric elastomer unit that switched operating modes. The host computer corrects the preset phase lead coefficient used by the periodic signal acquisition module based on the difference, so that the accuracy of the predicted arrival time gradually converges as the number of system runs increases.

Images