Tactile Displays Driven by Projected Light

Abstract

Tactile displays that lend tangible form to digital content could transform computing interactions. However, achieving the resolution, speed, and dynamic range needed for perceptual fidelity remains challenging. We present a tactile display that directly converts projected light into visible tactile patterns via a photomechanical surface populated with millimeter-scale optotactile pixels. The pixels transduce incident light into mechanical displacements through photostimulated thermal gas expansion, yielding millimeter scale displacements with response times of 2 to 100 milliseconds. Employing projected light for power transmission and addressing renders these displays highly scalable. We demonstrate optically driven displays with up to 1,511 addressable pixels -- several times more pixels than any prior tactile display attaining comparable performance. Perceptual studies confirm that these displays can reproduce diverse spatiotemporal tactile patterns with high fidelity. This research establishes a foundation for practical, versatile high-resolution tactile displays driven by light.

Figures (26)

• Figure 1: Fig. 1. Tactile display driven by projected light. (A) The display directly converts projected light into visible, tactile patterns via an array of optotactile pixels. (B) Light is converted to forces and displacements via an assembly of patterned layers, supporting manufacturing and scalability. From top: Elastic membrane (EcoFlex 00-10), cavity walls (polysiloxane, PS), thin photoabsorber (PGS, thickness: 17 $\mu$m), cavity walls (PS), optically transparent layer (acrylic). (C) Pixel operation (section view). An incident light pulse is converted into heat by the photoabsorber. Heat is transferred to air in the cavity, raising gas temperature $T_{gas}$ and pressure $P$. Gas expansion drives the deflection of the elastic membrane. (D) Finite element analysis of pixel thermo-mechanical response. (E) Flexible tactile display, 357 pixels. Scale bar: 1 cm. (F) Photostimulation elicits both tactile and optical feedback, due to the transmission of scattered light through the membrane. Extended exposure image of 1,511 pixel display. Scale bar: 1 cm. (G) This technique can be used to realize displays that vary in dimensions and pixel count. From upper left: 1, 8, 357, and 1,511 pixels. Scale bar: 1 cm. (H) Comparison with prior tactile displays (Fig. S15, Table S1). Our display enables rapid response rates, scales to pixel counts surpassing prior work, and is uniquely capable of both visual and tactile display. Driving mechanisms: L = light, EM = electromagnetic, PN = pneumatic, ES = electrostatic, T = thermal, PZ = piezoelectric, C = combustion, and EO = electroosmosis.
• Figure 2: Fig. 2. Thermal-mechanical response under photostimulation. (A,B) Absorber surface temperature and membrane deflection increased monotonically during illumination with power $P_L = 2.5$ W over 50 ms. (Scale bar = 0.5 mm). (C) Reducing the thermal bridge width from $w=$ 0.75 mm to 0.2 mm yielded higher temperatures and larger displacements. Scale bar: 0.5 mm. (D) Experimental absorber temperature $T(t)$ and membrane displacement $z(t)$ vs. time $t$ for $w=$ 0.2, 0.25, 0.4, 0.55, and 0.75 mm. As $w$ decreased, more heat was retained in the absorber, yielding higher temperatures and larger displacements. For $w=0.2$ mm, at $t=$ 50 ms we observed $T = 527$$^{\circ}$C and $d = 0.97$ mm. Grey dashed line: Numerical results for $w = 0.25$ mm; see Fig. S1. (E) The thermal resistance $R$ of the bridge increased as width $w$ decreased. The data agreed with the theoretically predicted relationship $R = a / w,$ with $a =$ 73.3 mm K/W, $r^2 = 0.98$. (F) Peak membrane displacement and absorber thermal relaxation time $\tau$ increased with thermal bridge resistance $R$, while the relaxation rate $1/\tau$ decreased, indicating that tuning $w$ involves a tradeoff between displacement and relaxation rate.
• Figure 3: Fig. 3. Cyclic photostimulated actuation. (A) Isometric force and unloaded displacement measurements. (B) Individual pixels produced peak forces of $F=$ 55 mN under photostimulation at power $P_L=2.5$ W and pulse duration $t_p=50$ ms (pulse energy 125 mJ). Force increased linearly with power $P_L$ and increased with pulse duration $t_p$. (C) Successive displacement responses to pulsed excitation exhibit minimal variation for pulse gap intervals $t_g \gg \tau$ (Fig. S7), where $\tau$ is the thermal relaxation time; Here, $\tau \approx 23$ ms and $R = 145$ K/W. At shorter gap intervals, $t_g,$ heat accumulates, yielding a slowly varying component, $d(t),$ and a reduction in the peak-to-peak oscillation amplitude, $\delta_{pp},$ relative to the first pulse amplitude, $d_1$. (D) As $t_g$ increases, oscillation amplitude, $\delta_{pp},$ (blue) increases; Regression fit: $\delta_{pp}/d_1 = 1-\exp{(t_g/\tau)}$ ($r^2 = 0.95$). The magnitude, $d$, of the slow component (red) followed an opposite trend, $d/d_1 = \tau/t_g$ ($r^2 = 0.81$). (Uncertainty bars: 1 SD, $n = 5$ trials). (E) Displacements $\delta_{pp}$ for pulse frequencies $f=$ 5 to 500 Hz. Peak-to-peak displacement decreased as $f$ increased, with $t_p/t_g = 1$. Insets: oscillating component, $\delta(t),$ for $f = 20$ Hz (yellow) and $200$ Hz (blue). (F) In sequential scanning, the maximum pixel display rate, $N,$ is dictated by the required displacement. At fixed $P_L$, $N$ decreases as displacement increases. For $P_L=2.5$ W, $N=$ 217 pixels/s is achieved at 50 $\mu$m, and $N=26.5$ pixels/s is achieved at 400 $\mu$m.
• Figure 4: Fig. 4. Perception of Dynamic Tactile Pattern Display. (A) Experiment 1: Perception of linear tactile motion patterns. All participants correctly reported the motion direction in 100% of trials. (B) Experiment 2: Perception of rotational motion for motion speeds $v=$ 16, 32, and 64 mm/s. Mean response accuracy: 94.7%. Black dots show the overall mean, and grey dots show participant mean. (C) Experiment 3: Spatial localization of actuated pixels near the finger. Mean accuracy: 78.4% (99% within 1 pixel distance). Mean localization error: 0.17 mm. (D) Experiment 4: The perceived intensity of pulse train feedback from a single pixel increased linearly with optical power $P_L$ ($\alpha = 0.022$, $\beta = -0.42$, $r^2 = 0.98$). Stimulus parameters: $N=20$ pulses, $t_p=t_g=$ 25 ms, $0.5 < P_L < 2.5$ W. (E) Experiment 5: Discrimination of temporal patterns. "Odd-One-Out" task to identify which of three sensations was different. Total accuracy: 89.6%. (F) Experiment 6: Perception of multi-point tactile patterns felt at any two of four fingers. Participants reported the locations that were stimulated with 96.3% accuracy.
• Figure 5: Fig. S1. Numerical Experiments A) The geometric and material setup for the 2D FEA simulation. B) Nearly identical mean cavity air temperatures were obtained using an effective heat transport model that accounted only for air conduction and using a full fluid dynamics simulation accounting for convection. C) Heat transport in the gas was primarily governed by conduction, with convection having a notable influence only within the first 1 to 2 milliseconds of heat application. D) Geometric and material setup for 3D pixel simulation. E) Snapshots of the FEA model at t = 5, 25 and 50 ms for $w = 0.4$ mm. The left color bar corresponds to the absorber temperature, whereas the right indicates membrane displacement. F) Time-resolved temperature and displacement curves for $w =$ 0.2, 0.25, 0.4, 0.55, and 0.75 mm. G) The membrane displacement decreases as absorber thickness, $h$, increases. H) The blocked force increases as the pixel radius increases.