Source-linked AI summary
Pneumatic Units for Logic-based Sequential Excitation (PULSE) in Wearable Haptic Devices
Jessica Healey, Anoush Sepehri, Michael T. Tolley, Tania K. Morimoto
TL;DR
Wearable pneumatic haptic devices offer lightweight, compliant feedback but usually require one valve and input per actuator, complicating portability. This paper introduces textile-based PULSE actuators with embedded fluidic logic and combines them in an optimized ring oscillator. The sleeve reduced required pneumatic inputs by 60% and rendered four directional cues for wrist guidance in a user study.
Problem
Wearable pneumatic haptic devices typically require a valve and input for each actuator, making fully portable systems challenging.
Method
The authors combine flat textile-based PULSE actuators with embedded fluidic logic into a ring oscillator and optimize design variables for target oscillation periods.
Results
The sleeve reduced required pneumatic inputs from five to two, a 60% reduction, and delivered four directional cues in a user study.
Takeaways & Limitations
PULSE embeds computation within soft actuators, expanding the portability of pneumatic haptic systems while supporting wearable tactile guidance.
Takeaways & Limitations
The system did not use an onboard air source, leaving technical details of full portability for future work.
Abstract
from arXiv · showhide
Soft, wearable robotic devices can deliver haptic feedback to support a wide range of tasks, such as extended reality, training various skills, and rehabilitation. Pneumatic actuation can deliver complex haptic feedback, is lightweight and compliant, and can be incorporated into textiles, making it promising for wearable applications. These soft pneumatic devices, however, typically require a valve and input for each pneumatic actuator, making it challenging to develop fully portable devices for at-home use. In this work we present a pneumatic unit for logic-based sequential excitation (PULSE). The PULSE is a flat, textile-based pneumatic actuator with embedded fluidic logic. By combining these actuators into a fluidic ring oscillator, we decreased the typical amount of required pneumatic inputs for a haptic forearm sleeve by 60%, with the ability to scale. We built the ring oscillator by optimizing design variables to reach desired periods of oscillation. We demonstrated a set of tactile stroking cues with periods ranging from 1.16 to 1.56 s and forces ranging from 1.07 to 2.04 N. We assessed the sleeve's ability to render differentiable, pleasant, and continuous haptic cues in a user study. The forearm sleeve containing PULSEs successfully delivered four directional cues and guided users to target wrist angles with fast reaction times, low overshoot amounts, and a 93.3% average accuracy of correct initial directions.
I. INTRODUCTION
Wearable pneumatic haptic devices are lightweight and compliant but typically require separate valves and inputs for each actuator, limiting portability. PULSE combines fluidic logic and tactile actuation in a flat textile-based unit, enabling scalable sequential cues with fewer pneumatic inputs.
- Pneumatic actuators provide lightweight, compliant haptic feedback and can render directional, continuous stroking, and social gesture cues.
- Separate valves for each pneumatic actuator create a large electronic backend and limit portability.
- PULSE combines pneumatic logic and haptic actuation in one soft, flat, textile-integrable unit.
- Five PULSEs sequentially inflate and deflate from one constant input and switch oscillation direction with a second input.
- Five valves are reduced to two, a 60% reduction, while the design can scale to any odd number of PULSEs with two inputs and valves.
II. MATERIALS AND METHODS
The PULSE is a textile-based actuator whose pressure-dependent inversion enables sequential operation in a pneumatic ring oscillator. The authors model oscillation timing using pneumatic resistances, capacitances, and buckling transitions, then update the model for observed resistance and pressure behavior.
- A. Actuator Working Principle: PULSE uses a single pneumatic input and a top tube that passes air when deflated but kinks and blocks flow when inflated.
- A. Actuator Working Principle: Embedding the actuator in fabric lets inflation produce normal force on the skin for tactile feedback.
- B. Modeling of Oscillation Period: In a ring, PULSEs use inverter behavior to generate sequential inflation from one constant input pressure.
- B. Modeling of Oscillation Period: The oscillation period is modeled from pneumatic tubing and chamber parameters, input pressure, and measured buckling and unbuckling pressures.
- B. Modeling of Oscillation Period: The updated model adds an additional pulldown-resistance term because Darcy-Weisbach resistance alone over-predicted periods as pulldown tubing length increased.
- B. Modeling of Oscillation Period: The total period combines each PULSE’s buckling and unbuckling times using effective resistances, PULSE capacitance, atmospheric pressure, and actuator count.
C. Expanding the Ring Oscillator
The expanded ring oscillator uses five PULSEs to produce sequential inflation and supports reversal of oscillation direction. A second tubing path and switch PULSEs enable the directional change while retaining shared pneumatic operation.
- Five PULSEs were selected to inflate sequentially along the forearm, using the odd actuator count required by ring oscillators.
- Connecting each output to a PULSE three positions farther along the line produced the desired sequential inflation.
- Shared supply pressure caused slight pressure fluctuations, but the resulting dip was not large enough for users to detect.
- A second tubing path arranged in the opposite order enabled the same PULSEs to inflate in the reverse direction.
- Two switch PULSEs cut off the unused tubing path to prevent air from leaking through the opposite direction.
D. Optimization for Desired Haptic Cues
The ring oscillator’s period depends on tunable physical design parameters and adjustable supply pressure. The authors optimized these variables to render continuous, pleasant, and effective directional tactile cues across a range of frequencies and forces.
- Physical design parameters can be tuned before fabrication, while supply pressure can be adjusted in real time to change the oscillator period.
- The design was optimized to achieve a range of oscillation periods, frequencies, and forces with one set of physical parameters.
1) Desired Haptic Cues:
The paper targets tactile cues that are pleasant, continuous, and clear for directional guidance, then optimizes PULSE geometry and tubing to produce them while limiting device bulk.
- Desired Haptic Cues: Periods near 1.2 s or 0.42 s produced the most pleasant and continuous stroking sensations in prior forearm-sleeve studies.
- Desired Haptic Cues: 1–10 cm/s stroking motions are reported as most pleasant for CT afferents in hairy skin, while pneumatic studies found highest pleasantness at higher speeds.
- Desired Haptic Cues: A period up to 1.5 s for a 16 cm stroke was predicted to provide clear, pleasant, and continuous sensations, with target forces beginning near 0.41 N.
- Design Optimization: The optimization maximized oscillation-period range across 40–140 kPa supply pressures while minimizing pulldown-resistor length.
- Design Optimization: The design imposed geometric constraints to reduce bulk, fit five PULSEs along the forearm, and maintain 40 mm center-to-center spacing.
E. Actuator Fabrication
PULSE fabrication combines heat-pressed textile layers, a PET-defined chamber, and a kinkable tube so the actuator both inflates and implements fluidic logic.
- Actuator Fabrication: Two TPU-coated Nylon sheets were heat-pressed around thin PET, creating an inflatable chamber and a heat-sealed section that buckles under pressure.
- Actuator Fabrication: The angled PET cutout determines the buckling geometry, and increasing its angle increases the pressure required to fold the fabric.
- Actuator Fabrication: 60° cutouts were selected because they combined a low buckling pressure with sufficient space for tubing attachment.
- Actuator Fabrication: Heat-shrink reinforcement and an indent localized the tube kink, helping the PULSE withstand external forces when integrated into the sleeve.
F. Forearm Sleeve Fabrication
The forearm sleeve uses layered textiles, pockets, and aligned straps to hold PULSEs snugly against the skin while increasing downward normal force.
- Forearm Sleeve Fabrication: Fabric layers form actuator pockets, and dimensions align with minimum average forearm measurements so tightened straps fit most forearms.
- Forearm Sleeve Fabrication: A Velcro strap aligned with each actuator produced the best fit on users.
- Forearm Sleeve Fabrication: A less-extensible top fabric layer than the bottom layer increases downward pressure on the forearm.
- Forearm Sleeve Fabrication: Each PULSE weighs 0.91 g, while the complete sleeve weighs 50 g.
III. RESULTS AND DISCUSSION
The predictive model agreed with experiments across key design variations, and the optimized final sleeve achieved the targeted oscillation periods and measurable haptic forces.
- Model Validation: 0.0611 s RMSE for three PULSEs and 0.229 s for five PULSEs quantified model agreement when changing actuator count.
- Model Validation: 2.50% error at 100 mm pulldown-resistor length was the best reported fit for that tubing-length comparison after adding the additional resistance term.
- Oscillation-Period Optimization: The optimized design predicted periods from 1.13 s to 1.50 s across a 0.370 s range.
C. Wearable Haptic Device Characterization
The PULSE sleeve rendered stroking cues across measured oscillation periods and forces, and was evaluated for comfort, cue effectiveness, and wrist guidance in a user study.
- Device characterization: 1.16 ± 0.03 s to 1.56 ± 0.01 s: the sleeve’s measured oscillation periods covered this range.The predicted and experimental periods had an RMSE of 0.035 s, corresponding to 2.12% error.
- Device characterization: 1.07 N to 2.04 N: PULSEs rendered forces across the tested supply pressures with a 0.5 N tension preload.This force range surpassed the minimum threshold and satisfied the 4% to 14% just-noticeable difference reported for similar forearm indentation tests.
- User study: 10 participants evaluated cue identification, pleasantness, continuity, and haptic guidance to target wrist positions.Participants’ vision and hearing were blocked during testing, and the protocol received institutional review board approval.
- Portable integration: The reduced pneumatic inputs supported integration with portable air sources, although this experiment used a compressed air line through the wall.The sleeve used a digital pressure regulator, microcontroller, and two solenoid valves to control its two inputs.
A. Cue Identification and Rankings
The sleeve conveyed four directional and intensity cues that users identified more reliably by direction than intensity, while ranking low-intensity cues as pleasant and continuous.
- Cue identification: 76.0%: participants’ average identification accuracy across all four haptic cues.Direction identification reached 94.5% for Up cues and 92.0% for Down cues, while intensity identification reached 81.5% for High and 78.5% for Low cues.
- Cue identification: 94.5% Up and 92.0% Down: directional interpretation exceeded the corresponding intensity discrimination.The four cues were Up/High, Up/Low, Down/High, and Down/Low.
- Pleasantness and continuity: 4.3 ± 0.67 pleasantness and 4.1 ± 0.74 continuity: users rated Low cues positively on the 5-point scale.High cues were also pleasant at 4.1 ± 0.88 but neutrally continuous at 2.8 ± 0.92; continuity differed significantly between High and Low cues, whereas pleasantness did not.
- Wrist guidance: Up cues signaled wrist flexion, Down cues signaled extension, and intensity indicated whether users were far from or close to the target angle.Feedback stopped after participants entered a 2.5° target deadband and held the target angle for three seconds.
- Wrist guidance: 93.3%: percentage of trials with a correct initial direction during target-angle guidance.Average maximum overshoot was 4.55 ± 7.88°, and the fraction of initially wrong directions was 0.0670 ± 0.117.
V. CONCLUSION
The paper presents PULSE as a flat, flexible pneumatic actuator that embeds logic within haptic actuation, reducing backend electronics for soft wearable devices.
- Future work: Robustness improvements and refined textile integration are identified as needed for PULSEs to be worn in everyday settings.The paper also leaves fully portable onboard air sources and further rehabilitation studies for future work.
- Conclusion: PULSE embeds fluidic logic in a flat, flexible haptic tactor, combining computation and actuation in one pneumatic unit.The design reduces backend electronics and is intended for integration into soft wearable devices.
- Conclusion: The authors characterized PULSEs in a ring oscillator, used a predictive model to optimize physical parameters, and evaluated comfort and guidance in a user study.The conclusion frames these evaluations as evidence for expanding the portability of pneumatic haptic systems.