Modular Pressure Redistribution Cushion with Proprioceptive Soft–Rigid Hybrid Actuator

Pressure ulcers, which can result from prolonged sitting, pose a significant challenge for wheelchair users. Soft robotics has considerable potential in preventing pressure ulcers. However, current soft robotics, constructed from flexible materials, face limitations including insufficient proprioception and controllability. Herein, a vacuum‐powered proprioceptive soft–rigid hybrid actuator (PSHA) module and a modular pressure redistribution cushion (MPRC) developed using this module are introduced. This PSHA module is capable of detecting both position and force. Each module within the MPRC is equipped with onboard control, proprioceptive sensation, and inter‐module communication. The MPRC incorporates a closed‐loop control system, enabling it to actively redistribute pressure, thereby preventing prolonged compression of local soft tissue during periods of inactivity. The proposed PSHA module, as evidenced in its application in pressure redistribution cushions, offers a promising approach for designs intent on reducing the risk of pressure ulcers. This study significantly contributes to the advancement of assistive technology, with the potential to enhance the quality of life for individuals with immobility or limited mobility.


Introduction
Pressure ulcers are localized damage affecting the skin and underlying tissue, resulting from prolonged pressure application on the skin. [1]Wheelchair users often spend the majority of their time sitting, putting them at risk for pressure ulcers. [2]These ulcers commonly occur in areas of bony prominence, such as the sacrum, ischial tuberosities, and the coccyx. [3]10] However, it has been observed that a majority of wheelchair users do not execute these weight redistributing movements with sufficient frequency. [11]External methods aiming to diminish the magnitude and duration of pressure are imperative for mitigating the susceptibility to ulcer formation.
[14] In recent years, numerous cushions for redistributing pressure have been developed, which can be classified into passive and active pressure redistribution strategies. [15,16]Passive pressure redistribution cushions utilize a variety of features and materials, including foam, gel, air cells, or a combination thereof, to reduce localized pressure points by evenly distributing the individual's body weight over a larger surface area. [17]Passive pressure redistribution cushions may not offer optimal relief for all users, given their inherent inability to actively adapt to the unique contours, weight distribution, and movements of individual bodies.Furthermore, individuals with specific pressure-sensitive areas may find that passive cushions lack the capability to offer adjustable or dynamic pressure relief, thus potentially limiting their effectiveness in addressing localized pressure points.Active pressure redistribution cushions can dynamically adjust their shape, pressure distribution, and position in response to an individual's movements, weight, and body contours.Active pressure redistribution cushions commonly employ air or fluid-filled chambers, [18] adjustable cells, [19] or electronic controls [20] to offer customizable and dynamic pressure relief.[23][24] The alternating pressure mechanism may induce a perception of fluctuating pressure or displacement, potentially exacerbating pressure on susceptible or preexisting damaged regions.
[27][28] Recent advancements in soft robotics technologies demonstrate significant potential for the development of pressure redistribution cushions.Low et al. proposed an innovative approach involving the integration of soft sensors and actuators within conventional insoles, aiming to redistribute plantar pressures effectively and thereby mitigate risks associated with excessive pressure concentrations. [29]Raeisinezhad et al. developed a modular, intelligent soft robotic pad (IntelliPad) for pressure injury prevention. [30]Robertson et al. designed a soft, reconfigurable robotic surface that exhibited a notable ability to reduce and redistribute the forces applied on the surface, thereby offering an efficient approach to pressure distribution. [31]Deng et al. proposed a soft robotic table, composed of multiple soft actuators, which can generate surface deformation by chamber inflations and deflations. [32]Takashima introduced an advanced active air mattress, comprising an array of soft air cells, and integrated with a sensitive pressure sensor sheet.This setup allows for independent control of each air cell, thereby offering a highly adaptable pressure distribution solution. [33]Despite the benefits of these existing solutions, most of the aforementioned soft actuators lack a crucial feature in the development of active pressure redistribution cushions: the accurate detection of contact force.
To manage interface pressure effectively in an active cushion, several crucial factors must be considered, including pressure distribution, peak and mean pressure values. [34]A well-designed cushion should not only accurately sense pressure, but also distribute it uniformly across the contact area.This prevents the excessive concentration of interface pressure on specific points, which can lead to discomfort or tissue damage.
Here, we propose a vacuum-powered proprioceptive soft-rigid hybrid actuator (PSHA) module.This module features a unique hybrid design, combining a flexible soft skin with a robust rigid skeleton.The soft skin provides flexibility and safety due to its inherent compliance, while the rigid skeleton undergirds the structure, enhancing its durability and precision.This module incorporates a time-of-flight (ToF) ranging sensor and a barometric sensor, both integrated into the bottom surface of the vacuum chamber, to provide feedback information.The ToF ranging sensors utilize the duration taken by photons to travel between two points to calculate the distance separating these points.The performance of the ToF sensor is described in Supporting Information, along with Figure S1, Supporting Information.Employing the ToF ranging sensor, we can quantify the positional variations of the PSHA module.By measuring position changes and air pressure, the contact force exerted by the PSHA module can be calculated through a closed-loop model (detailed in Section 2).Therefore, the motion of the PSHA module can be easily controlled by the input air pressure.
By integrating multiple PSHA modules, we introduce a novel modular pressure redistribution cushion (MPRC) equipped with both contact force and configuration sensing capabilities.Utilizing a closed-loop control system, the MPRC has the potential to alleviate interface pressure and eradicate peak pressure points under a range of circumstances (Figure 1a,b).The modular nature of the MPRC facilitates the easy addition or replacement of individual PSHA modules, thereby enhancing the reliability and maintainability of the entire system.
The main contributions of this work are as follows: 1) a vacuum-powered PSHA module with proprioception of position and contact force is proposed comprehensively with design, fabrication, and modeling; 2) an MPRC is developed for the prevention of pressure ulcers.The proposed MPRC design demonstrates a significant advancement in pressure management technology, with the unique capability to efficaciously reduce peak and average pressure values.
The rest of this article is organized as follows.Section 2 provides an overview of the design, fabrication, modeling, and control of the PSHA module.Section 3 presents the concept, design, and control of the MPRC.Section 4 describes the experimental validation of the cushion, including assessments of proprioception and pressure redistribution.Section 5 discusses the characteristics of the MPRC.Finally, Section 6 summarizes the conclusions of this study and provides directions for future research.

Design and Fabrication of PSHA
The fabrication and assembly of the proposed PSHA module prototype (Movie S1, Supporting Information), including details of specific components and their arrangement, are outlined in Figure 1c,d.The proposed PSHA module comprises a soft skin and a rigid skeleton.The soft skin, which is made from thermoplastic polyurethane (TPU), has a thickness of 0.3 mm.The rigid skeleton employs a compressible metal spring.The TPU film is shaped into a cylindrical form that envelopes the spring, with the top and bottom sealed using end caps.The caps are 3D printed by photocurable resin.The caps are interconnected with a pair of ropes that limit the extension of the spring and measure 100 mm in length.The two limiting ropes restrict the active length of the PSHA module to 100 mm.The metal spring, which has an initial free length of 150 mm, is placed between the top and bottom caps.This arrangement imparts a compression of 50 mm to the spring.
To obtain the position and internal air pressure information of the PSHA module chamber, we employ a ToF ranging sensor (VL53L0X, STMicroelectronics) and a barometric sensor (LPS22HB, STMicroelectronics), both of which are centrally positioned on the chamber's base, as illustrated in Figure 1d-f.The PSHA module is affixed to a mounting base, which houses an electronic control board for sensor data acquisition.This electronic control board, consisting of an STM32 minimum system board and a communication board (MAX485), contains embedded electronics that are specifically designed to acquire and process the sensor information obtained from the PSHA module, ensuring precise and dependable control.The mounting base provides a stable platform that simplifies the installation and integration of the PSHA module into larger systems.

Modeling of PSHA
The PSHA module operates through a vacuum-powered mechanism.The force produced by vacuum pressure within the module is exerted on the rigid skeleton.The behavior of the PSHA module is characterized by a quasi-static model.We assume that the soft skin of the PSHA module is highly deformable, inextensible, and has no stiffness.Based on the principles of virtual work, the force generated by the vacuum pressure is dependent on the length of the PSHA module under a consistent air pressure difference, and can be expressed as where F a is the force generated by the vacuum pressure, P 0 app is the inner air pressure of the PSHA chamber, dL is the change in length of the spring pitch, and dV a is the change in volume of the PSHA module chamber.
The space within the PSHA module chamber is divided into three segments: the interstitial region within the active coils of the spring (Figure 2a), the space between the bottom cap and active coils of the spring (Figure 2b), and the region between the top cap and the active coils of the spring.Moreover, it is assumed that the spring diameter remains constant, consequently maintaining a constant cross-sectional area of the PSHA module throughout the duration of the contraction.The volume V act of chamber within the active coils of the spring can be approximated as a cylinder, in which the curved surface folds inward, forming a torus shape as depicted in Figure 2a.The mathematical expression is as follows.
where R is the radius of the PSHA module, θ is half of the central angle of the circular segment of the curved surface, and r is the radius of curvature of the skin, and it can be expressed as where L 0 is the initial length of the spring pitch.The inextensibility of the PSHA module makes the length L of the spring pitch as a function of θ, and it can be written as [35] L The volume V bot between the bottom cap and active coils of the spring can be calculated as The volume V top between the top cap and the active coils of the spring is equal to V bot .The PSHA module volume can be calculated as Substituting Equation ( 5) and ( 7) into Equation ( 2) and the result gives 38] As the film begin to stick to the caps (Figure 2c), the effective area A e decreases (Figure 2d).This can be calculated geometrically as follows The force F a can be calculated as The force F a generated by the applied vacuum pressure is calculated using Equation ( 8) for contraction ratios below 36.34%, and using Equation (10) for contraction ratios above 36.34%.
A compression test was performed to determine the structural elastic force F e of the PSHA module using a hand-operated mechanical testing machine.To improve the accuracy and reduce the computational complexity, a quadratic fitting is used to model the relation between force and contraction ratio (Figure S2, Supporting Information).The elastic force F e can be expressed as When the PSHA module is at the equilibrium state, the vacuum pressure force F a , structual elastic force F e , and contact force F c achieve an overall force equilibrium, as illustrated in Figure 2e.The contact force F c can be expressed as Further information about the derivation of Equation ( 3), ( 6), (8), and ( 9) is included in Supporting Information, along with Equation (S1)-(S13), Supporting Information.

Force and Position Control
Once position and force-related data have been acquired, the performance of the PSHA module can be effectively regulated through input air pressure.Instead of using costly proportional or servo valves as the control elements of the vacuum system, we employ high-speed electromagnetic valves.High-speed electromagnetic valves allow for swift and accurate manipulation of air inflow and outflow, making the control system more economical and effective.The operational diagram illustrating the control of position and force in the PSHA module is presented in Figure 3.
A miniature vacuum pump (voltage: 12 V DC, maximum flow rate: 15 LPM) is used as the vacuum pressure source actuating the PSHA module.The vacuum pressure generated by the vacuum pump is channeled to the sealed chamber inside the PSHA module through an electromagnetic valve matrix.Two high-speed electromagnetic valves (T10, OST Pneumatic) regulate air inflow and outflow for the sealed chamber within the PSHA module.Connected through a T-fitting to a single tube extending to the PSHA module, one valve controls vacuum pressure while the other manages atmospheric pressure.The valves are controlled by pulse width modulation (PWM) signals.Owing to the shared tube, only one of the valves is operated at a time.In addition, the microcontroller, along with metal-oxide-semiconductor field-effect transistors (MOSFETs), governs the operation of the solenoid valve, which in turn influences the state of the PSHA module.
The barometric sensor and ToF sensor in the PSHA module provide real-time feedback to the microcontroller unit, enabling precise adjustments to the PWM signal.The feedback is processed by a proportional-integral-derivative (PID) controller, which adjusts the duty cycle of the PWM signal to achieve the desired air pressure conditions within the PSHA module.The polarity of the controller's output indicates either air inflation or exhaust, offering fine-tuned control over the module's air pressure conditions.
The PSHA module operates in two distinct modes: position control and contact force control.In position control mode, the difference between the expected position and the measured feedback position from the ToF ranging sensor is used as the input for the position PID controller, which runs on the microcontroller.
Similarly, in contact force control mode, the difference between the expected contact force and the force computed by the aforementioned numerical model is used as the input for the force PID controller, which also runs on the microcontroller.

Design of MPRC
A novel design concept for an MPRC is introduced that merges both active and passive components to improve pressure management.As illustrated in Figure 4a, the active region of the cushion design consists of a grid configuration containing 25 PSHA module units arranged in a 5 Â 5 layout.This deliberate arrangement facilitates the efficient redistribution of pressure, thereby ensuring optimal user comfort.The passive area incorporates sponge of a carefully selected density of 30 kg m À3 .The integration of high-density materials provides the cushion with structural stability, contributing to its durability and prolonged performance.The overall dimensions of the cushion are 450 Â 450 Â 150 mm, as illustrated in Figure 4b,e.

Control System
The MPRC design leverages modularity and distributed control principles.This modular methodology decomposes the system into discrete, functional entities that offer seamless integration and interchangeability, thereby promoting system adaptability, flexibility, and ease of maintenance and repair.
The cushion's pneumatic design, as depicted in Figure 4c, follows a modular approach.The previously mentioned miniature vacuum pump serves as the source of vacuum pressure input, which is distributed among all the PSHA modules within the MPRC.To streamline control over these modules, a pneumatic circuit encompassing 50 solenoid valves was assembled.Within the cushion, each PSHA module was interconnected with two high-speed electromagnetic valves.As mentioned before, the first valve was responsible for adjusting the vacuum pressure, while the second valve was utilized to regulate the atmospheric pressure.The two valves can be controlled by the electronic control board (see Section 2.3), which is embedded in the PSHA module's mounting base.
The electrical system is designed to manage the valves of each individual PSHA module.The system, illustrated in Figure 4d,e, comprises a 12 V adapter for energizing the solenoid valves, 50 MOSFETs for maneuvering the valves, and a 12-5 V converter supplying power to the electronic control boards of the PSHA modules.Within the pressure redistribution cushion, every PSHA module comes outfitted with its dedicated electronic control board.This control board is composed of an STM32 minimum system board and a communication board.The control board processes real-time feedback from the ToF ranging sensor and the barometric sensor.These control boards are connected in series using power and ground lines, as well as an RS485 communication bus.Each control board has a unique ID.The boards are ultimately connected in a daisy chain configuration, managed by a personal computer (PC) connected at the start of the chain, as shown in Figure 4c.
A customized communication protocol is developed to ensure efficient communication between the PC and the control boards.The protocol is specifically designed to transmit commands, retrieve PSHA module status updates, and prevent packet collisions. [39]The communication package is composed of 18 bytes (Bytes 0-17).Each byte has a specific role within the communication protocol.The allocation of these bytes is as follows: Byte 0 represents the start flag of the communication package; Byte 1 is reserved for the control board ID; Byte 2 indicates the command mode, indicating the type of command being transmitted; Bytes 3-13 contain the content of the package; Bytes 14-16 comprise the cyclic redundancy check frame for error detection and verification; and Byte 17 represents the stop flag of the communication package.This structured arrangement ensures organized and reliable data transmission between the PC and the control boards, promoting effective communication within the system.The command modes for the PSHA modules are divided into three categories: motion execution, parameter updates, and module status monitoring.Communication between the control boards and the PC is facilitated via an RS485 communication bus, operating at a baud rate of 115 200 bps and exhibiting a communication delay of 5 ms.Control boards interpret packets transmitted by the PC through the serial port, executing commands contained within packets that correspond with their respective IDs.These commands can encompass PID parameter updates, establishing new target positions for the PSHA module, determining target contact forces, and initiating requests for updates on the PSHA module's status information.When the PC solicits a data response, usually pertaining to the PSHA module's status information, the corresponding control board transmits the necessary data via the communication bus.
We designed a graphical user interface (GUI) using MATLAB APP Designer (Figure S3 and Movie S2, Supporting Information).This GUI provides a visual representation of the contact force and position of each PSHA module within the cushion.It also displays the overall body pressure distribution, providing a comprehensive view of system performance.The GUI enhances user interaction and allows for real-time monitoring and control of the PSHA module behavior.
The primary aim of this study is to control the interface pressure effectively between the user's hip and the cushion by regulating the contact force of each PSHA module within the cushion.To achieve this, the PSHA modules are operated in contact force control mode, with a predefined safe contact force threshold assigned to each PSHA module.When a load (body weight) is applied to the cushion surface and if the force becomes concentrated on a single module exceeding the defined threshold, the resulting contact force concentrated at that module becomes significant.The force is then redistributed among the adjacent modules when the load at the specific concentration point is reduced.In this study, the force threshold value is set to 17 N (equivalent to a pressure of 33 mmHg).This threshold value is deliberately set below the critical level of 35 mmHg, which has been identified as the threshold for inducing pressure sores. [40]he contact force exerted by each PSHA module is computed by its control board.To control the contact force for each PSHA module, individual PID controllers are implemented on the control boards, enabling autonomous operation of all PSHA modules.Should the contact force exceed the established threshold, the control board adjusts the internal chamber air pressure of the corresponding PSHA module to decrease the contact force.

Model Validation
To evaluate the accuracy and performance of the developed model, we conducted a series of quasi-static experiments.In our investigation, we acquired the real-time position of the PSHA module using a ToF ranging sensor and measured the real-time chamber air pressure using a barometric sensor.For the PSHA module, the TPU film's extensibility incurred force losses.Higher vacuum pressure caused the film to elongate more, leading to discrepancies between experimental forces and model predictions because of these force losses.Based on the experiment data in Figure 5, we introduced a correction factor C to the model, and it can be expressed as C ¼ À1.397 Â 10 À6 Â P 0 app (13)   Consequently, the comprehensive force equation incorporating the proposed correction factor is as follows.
The contact force prediction capability of the proposed model for the PSHA module was tested.The experimental platform is presented in Figure S4, Supporting Information.A series of quasi-static compressive tests of the PSHA were conducted using a hand-operated mechanical testing machine.During the tests, we subjected the PSHA module to passive compression under various chamber vacuum pressures, specifically 2.5, 5, 7.5, 10, 12.5, and 15 kPa.In accordance with our observation that an increased pressure differential results in a more pronounced deformation of the PSHA, we decided not to conduct further tests under higher vacuum pressures.The test outcomes were then compared with the force predictions inferred by the analytical model, as shown in Figure 5a.The model proposed for the PSHA module successfully delineated the overall trend of the experimental results and precisely predicted the axial force of the PSHA module with a mean error 0.6 N.This agreement confirms the robustness of our model and its predictive accuracy in describing the PSHA's mechanical behavior.The discrepancies observed during comparison with the experimental data may be attributable to variations in the deformation patterns of the TPU film during contraction.
Under the influence of vacuum pressure, the PSHA module exhibited a tendency to contract, thereby generating a measurable output force.This output force generated by vacuum pressure was quantified using the experimental setup outlined in Figure S5, Supporting Information.As illustrated by the isobaric curves presented in Figure 5b, the established model predicted the output force with a mean error of 0.8 N, further affirming the precision of the constructed model.
Experiments were conducted to investigate the relationship between the contraction ratio and vacuum pressure within the PSHA module.To create the required vacuum conditions, a vacuum pump was employed, with the output vacuum pressure regulated by adjusting the setting on a regulator (AFC2000, CHNT).The vacuum pressure in the PSHA chamber was incrementally increased from 0 to 50 kPa.The relationship between contraction ratio and vacuum pressure is illustrated in Figure 5c.The prediction error was approximately 7% during the period of maximum free contraction, which was observed at a vacuum pressure of 48 kPa.The mathematical model accurately captured the correlation between vacuum pressure and contraction ratio.This correlation was especially pronounced in systems driven pneumatically at low vacuum pressures.

Performance of PSHA
In this study, we conducted a step signal tracking test to evaluate the PSHA module's performance in terms of position and force control.For the position control assessment, a comparative analysis was conducted between the command position and the feedback position obtained from the ToF sensor housed within the PSHA module.The results are presented in Figure 6a, showcasing effective linear position control accuracy within AE1 mm.For the force control assessment, the feedback force, derived based on the prescribed model, was contrasted with the command force; these results are depicted in Figure 6b, demonstrating a force control accuracy within AE2.2 N.These findings suggested that the PSHA module exhibited exceptional precision in managing both position and force, highlighting its potential as a reliable component for diverse applications.
We conducted life testing on the PSHA module.The module, driven by vacuum pressure, moved back and forth between positions of 90 and 75 mm for a total of 600 cycles.Figure 6c,d illustrates the variations in position and barometric value of the module.Notably, the PSHA module remained airtight and stable throughout the range of movements.Nonetheless, as we repeated the cycles, a decrease in the requisite vacuum pressure to achieve the stipulated position was observed, possibly because the TPU film stretched more with each cycle.

Performance of MPRC
Having verified the position, force sensing, and control capabilities of the PSHA module, we performed extensive pressure redistribution assessments to appraise the MPRC's functionality for active pressure management.A closed-loop force control was instituted for each PSHA module to facilitate active adjustments of the interface pressure between the MPRC and an individual.We conducted a comprehensive pressure redistribution test to evaluate the viability of the MPRC for active pressure management.To assess the efficacy of the automated pressure redistribution function, we utilized a commercial interface pressure sensor mat (Model 5330, Tekscan), which was placed on the MPRC to measure the interface pressure distribution.We enlisted an able-bodied subject (male; 31 years old; height: 165 cm; mass: 56 kg).The subject was a volunteer (first author of the manuscript) and provided informed consent.The study received approval from the Ethics Committee of Beihang University, School of Biological Science and Medical Engineering (Approval Number: BM20200055).
Before initiating the experiment, we calibrated all the sensors inside the PSHA modules.We operated the solenoid valve that manages the atmosphere, allowing the internal chamber of the PSHA modules to equilibrate with the external environment.At this stage, the pressure inside all the PSHA modules equalized with atmospheric pressure, and the modules were adjusted to their initial free positions.Concurrently, the individual control boards calibrated the ToF sensors and barometric pressure sensors within their respective modules.Following these preparations, we instructed the subject to sit on the MPRC with his feet suspended off the ground.
An initial measurement was taken with the force control deactivated while the subject adopted various postures, including upright, left-leaning, right-leaning, backward, and forwardsitting postures.Additional measurement was taken after the force control was activated.Concurrently, during the pressure redistribution test, the displacement of the PSHA modules nested within the MPRC was monitored and recorded as a measure of immersion.Data pertaining to interface pressure, captured by the sensor mat, alongside the applied contact force on the PSHA modules within the MPRC and the contraction ratio of the PSHA modules pre-and post-force control activation are illustrated in Figure 7a-c.The depicted scenarios encompass upright, left-leaning, right-leaning, backward, and forwardsitting postures.Following the activation of the force control, the modules that were subjected to forces exceeding a threshold value of 17 N actively contracted, which in turn increased the contact area and decreased the interface pressure between the MPRC and the subject.This process facilitated a significant reduction in the maximum interface pressure, as recorded by the sensor mat, indicating reductions of 32.2%, 52.6%, 52.7%, 50.5%, and 54%, respectively, for each posture.Concurrently, the average interface pressure also decreased, exhibiting reductions of 7%, 23.9%, 32.3%, 27.7%, and 15.1%, respectively, for each posture.A comparative analysis of the data before and after force control activation revealed a significant decline in both peak and average interface pressure metrics across all evaluated seating postures.
Figure 8 and Movie S2, Supporting Information illustrate the results of continuous recordings (sampling rate of 1 Hz) of the maximum contact force, maximum gradient force, mean force, and mean contraction of all PSHA modules within the MPRC.These data substantiated the capability of the MPRC to dynamically modify both its shape and the distribution of contact forces in real time in response to the user's physical activities, thereby systematically redistributing contact pressure to mitigate the potential onset of pressure ulcers.
Within the MPRC, each PSHA module can modify its operating mode to meet user-specific requirements.More specifically, the position control mode of the PSHA can be utilized to alleviate pressure in targeted skin areas.To validate the pressure offloading capability of the MPRC, we conducted a test focused on the left ischial tuberosity, utilizing the same setup as in the preceding pressure redistribution test.During this experiment, we initiated the operation by activating the module located at the position corresponding to the subject's left ischial tuberosity, setting it to the position control mode with a target position of 65 mm to lessen the force applied to that area.Subsequently, the remaining PSHA modules within the MPRC, operating in force control mode, aided in redistributing the pressure.Data collected-including interface pressure from the sensor mat, the contact force applied to the PSHA modules in the MPRC, and the contraction ratio of these modules during the testare illustrated in Figure 9a-c.The results demonstrate a significant reduction in the average interface pressure at the left ischial tuberosity, amounting to a 60% decrease when both force control and position control are activated concurrently.This suggests that the MPRC can effectively relieve pressure on specific skin areas without compromising its overall pressure redistribution capabilities, potentially making it a suitable option for patients with localized skin sensitivity.

Discussion
The MPRC possesses the unique ability to dynamically adapt its shape and pressure distribution in response to the user's movements, weight, and body contours, an attribute facilitated by its modular design approach and distributed control scheme.Each PSHA module within the MPRC functions autonomously, overseen by its individual control board, thus preserving independent functionality.This architecture guarantees that the addition of more modules does not compromise the system's responsiveness or its efficacy in redistributing pressure.A primary limitation in increasing the number of PSHA modules is the confined dimensions of the seat cushion.To accommodate more modules within the constrained space of the cushion, a reduction in the diameter of the PSHA module is necessary.The proposed module, with a diameter of 70 mm, presents a parameter that can be adjusted relatively easily.However, it is vital to note that reducing the diameter of the spring skeleton within the PSHA module might increase the stiffness of the metal spring, potentially affecting the MPRC's efficiency in redistributing pressure.
In comparison to previously documented pressure redistribution cushions employed for pressure ulcer prevention, we contend that our presented MPRC provides several substantial advantages.While most active cushion systems are constructed using soft materials-resulting in a lack of precise models to describe their behavior and a decreased ability to sense contact forces accurately-our design alleviates these challenges.We have formulated quasi-static models for the PSHA module, enabling precise control of both position and force.This advancement equips the MPRC with a heightened capability to sense and modulate contact forces, thus decreasing control complexity and response time-a notable enhancement compared to the systems detailed in ref. [17].Furthermore, the modular concept central to the MPRC's design ensures that damage to a single module (like air leakage) won't affect the performance of the other modules.The impaired module can be replaced with ease, circumventing the need to revamp the entire system.This modular strategy thus augments fault tolerance and simplifies maintenance of the MPRC, positioning it as a preferable alternative to alternating pressure cushions as discussed in refs.[18,19,24], and marking the MPRC as a user-friendly option.

Conclusions and Future Work Direction
In this study, we introduce a novel MPRC based on a new type PSHA module, which is capable of operating in both position and force control modes.Notably, the PSHA module we developed can sense and control both the position and contact force.Design, fabrication, and modeling of the PSHA are presented in detail.Unlike traditional methods that employ resistive [41][42][43] or capacitive [44][45][46] measurements from deformable sensing components to gauge interface pressure variations, the PSHA module uses a ToF optical sensor for contact force measurement.This noncontact method of measurement enhances the equipment's durability and lifespan.A quasi-static model is developed to explain how the PSHA module is actuated pneumatically.Dedicated experiments are conducted to validate the basic  performance of the PSHA module with integrated actuation, execution, and proprioception capabilities.
The MPRC is developed by integrating multiple PSHA modules, thereby enabling the integration of contact force and configuration sensing capabilities.The use of both soft and rigid materials in the PSHA supports its modular design, which facilitates the convenient integration of hardware components and circuit boards.The modular design approach of the MPRC offers several benefits, including flexibility, scalability, fault tolerance, rapid prototyping, and customization.Each PSHA module is equipped with onboard control, proprioceptive sensation, and inter-module communication.Closed-loop force control is utilized to actively modulate the interface pressure between the cushion and the seated individual.Experimental results demonstrate that the proposed MPRC can identify and automatically redistribute pressure away from high-pressure points.
This study provides a novel perspective on pressure redistribution cushion design.The corresponding modeling methods and manufacturing processes are introduced.However, certain noteworthy limitations are evident in the practical application, primarily due to the dimensions of the cushion and the size of the control hardware.The cushion's modular framework facilitates convenient customization to accommodate a variety of user needs and operational contexts.Anticipated advancements, such as the integration of more compact solenoid valves and MOSFETs within the PSHA module's mounting base, are expected to significantly reduce the control box's footprint.Future research will focus on minimizing both the cushion and control box dimensions, thereby facilitating seamless wheelchair integration and enhancing user comfort through PSHA module optimization.Moreover, efforts will be geared toward the development of a high-resolution module layout.In addition, we aspire to refine the control algorithm to proficiently identify user movements, thereby enabling precise body positioning and stabilization.Furthermore, we plan to extend clinical trials for comprehensive validation with a broader group of patients.

Figure 1 .
Figure 1.a) The structure of the modular pressure redistribution cushion (MPRC).b) Schematic illustration of the MPRC and pressure redistribution process.c) Fabrication and assembly process of the proprioceptive soft-rigid hybrid actuator (PSHA) module.d) Fabricated PSHA module.e) Time-offlight (ToF) ranging sensor and barometric sensor.f ) Control board.

Figure 2 .
Figure 2. a) Volume (V act ) of chamber inside the active coils of the spring.b) Volume (V bot ) of the chamber located between the bottom cap and the active coils of the spring.c) Side view of the PSHA module once the contraction ratio exceeds 36.34%.d) Effective area of the PSHA module when the contraction ratio exceeds 36.34%.e) The force balancing diagram for the PSHA module.

Figure 3 .
Figure 3. Schematic of closed-loop proportional-integral-derivative (PID) position control and contact force control, using high-speed solenoid valves.

Figure 4 .
Figure 4. a) The components of the MPRC.b) Dimensions of the MPRC.c) Illustration of the MPRC's electrical and pneumatic systems.d) Photograph of the control box.e) Frontal view of the MPRC and control box.

Figure 5 .
Figure 5. a) A comparison between the model predictions and experimental results concerning the compression tests at varying vacuum pressures (2.5, 5, 7.5, 10, 12.5, and 15 kPa).b) A display of the isobaric curves associated with the PSHA module.c) A comparison between the model predictions and experimental results regarding the contraction ratio of the PSHA module under different vacuum pressures.

Figure 6 .
Figure 6.a) Comparison between the reference values and experimental results for position control.b) Comparison between the reference values and experimental results for force control.c) Variations in the position of the PSHA module during the life test.d) Variations in the barometric values of the PSHA module during the life test.

Figure 7 .
Figure 7. a) Interface pressure maps before and after force control activation across various sitting postures.b) Contact force maps of PSHA modules before and after force control activation across various sitting postures.c) Contraction maps of PSHA modules before and after force control activation across various sitting postures.

Figure 8 .
Figure 8. Results of continuous recordings of the maximum contact force, maximum gradient force, mean force, and mean contraction of all PSHA modules within MPRC during continuous adjustments of the user's sitting posture.