LumiTac: Compact, High-Rate Six-Axis Optical Tactile Sensing for Human-Robot Interaction

Hao Wu; Haotian Guo; Yu Feng; Yutong Wang; Jianshu Zhou

cs.RONov 4, 2026

LumiTac: Compact, High-Rate Six-Axis Optical Tactile Sensing for Human-Robot Interaction

Authors: Hao Wu, Haotian Guo, Yu Feng, Yutong Wang, Jianshu Zhou

Organizations: Department of Mechanical Engineering, National University of Singapore, Singapore

Abstract

Integrating six-axis tactile feedback into robotic interfaces requires compact sensors with accurate load measurement and rapid online reconstruction. This paper presents LumiTac, a low-profile, camera-free optoelectronic tactile sensor combining spatially distributed illumination with spectral encoding. Five photoemitters and five RGBW photoreceivers are mounted on opposing circuit boards coupled through translucent silicone. Load-induced interlayer motion modulates the internal light field under compression, shear, bending, and torsion. The sensing head measures 12 mm in diameter and 4.75 mm in thickness, weighs less than 2 g, and has a component cost of approximately USD 7. A compact multilayer perceptron reconstructs three force and three moment components from 20 optical channels. Independent testing yielded coefficients of determination (R-squared) of 0.9657-0.9941, with mean absolute errors of 0.054-0.190 N for forces and 0.304-1.419 mN m for moments. Online prediction achieved an average update rate of 303.57 Hz and a median software processing latency of 0.722 ms. Experiments involving combined loading, light weights, and rapid tapping evaluated load tracking, sensitivity to small forces, and transient response, respectively. Robotic demonstrations further validated its applicability to tactile-guided manipulation and physical human-robot interaction. This study paves the way for compact, high-rate optical tactile sensing in robotic manipulation and human-robot interaction.

Figures & tables

Fig. 1: LumiTac prototype and potential applications. (a) Size comparison with a coin; (b) side view of the sensor structure; (c) application concepts for robotic grippers and dexterous hands.

Sensor	Transduction	Size [mm]	Mass (g)	Reported load ranges $F$ : N; $M$ : mN $\cdot$ m	Accuracy $F$ : N; $M$ : mN $\cdot$ m	Sampling rate	Cost (USD)
LiVec finger [ 7 ]	Optical	$26.44\times 38.22$ $\times 12$	NR	$F$ : $(\pm 2,\pm 2,0\text{--}12)$ $M$ : NR	Error SD $F$ : $(0.112,0.092,0.139)$ $M$ : $(1.90,1.54,1.26)$	390 Hz	NR
El-Azizi et al. [ 9 ]	Optical	$\phi 27\times 20.82$	16.4	$F$ : $\sim(\pm 5,\pm 5,0\text{--}5)$ $M$ : $\sim(\pm 200,\pm 200,\pm 200)$	MAE $F$ : $(0.060,0.050,0.070)$ $M$ : $(2.62,2.68,0.64)$	$>2.5$ kHz	100
F-TOUCH [ 10 ]	Visuotactile	NR	NR	$F$ : $(\pm 1,\pm 1,0\text{--}8)$ $M$ : $(\pm 20,\pm 20,\pm 20)$	RMSE $F$ : $(0.018,0.016,0.094)$ $M$ : $(0.657,1.566,0.407)$	30 Hz	NR
Mao et al. [ 6 ]	Thermal	$12\times 15\times 5$	0.30	$F$ : $(\pm 5,\pm 5,0\text{--}15)$ $M$ : $(\pm 60,\pm 60,\pm 20)$	RMSE $F$ : $(0.180,0.150,0.180)$ $M$ : $(0.58,0.62,0.26)$	NR	NR
CoinFT [ 11 ]	Capacitive	$\phi 20\times 2$	2.0	$F$ : $(\pm 5,\pm 5,0\text{--}15)$ $M$ : $(\pm 100,\pm 100,\pm 80)$	RMSE $F$ : $(0.186,0.134,0.186)$ $M$ : $(1.114,1.322,0.810)$	360 Hz	$<11$
HexaTouch [ 12 ]	Capacitive	$15\times 15\times 8$	NR	$F$ : $(\pm 5,\pm 5,0\text{--}25)$ $M$ : $(\pm 100,\pm 100,\pm 50)$	RMSE $F$ : $(0.140,0.130,0.130)$ $M$ : $(2.48,2.50,0.95)$	NR	NR

TABLE I: Comparison of representative multiaxis tactile and force/torque sensors.

Fig. 2: Structure and sensing principle of LumiTac. (a) Layered structure of the sensor; (b) assembled sensor and arrangement of light-emitting elements and receivers; (c) cross-sectional view of the internal optical paths; (d) 20-channel RGBW signal readout and MLP-based six-axis force and moment reconstruction; (e) schematic illustration of optical path changes under normal, shear, and torsional loading.

Fig. 3: Simulated deformation and optical responses of LumiTac under single-axis loading. (a)–(c) Responses to $F_{x}$ , $F_{y}$ , and $F_{z}$ , respectively; (d)–(f) responses to $M_{x}$ , $M_{y}$ , and $M_{z}$ , respectively. Each panel shows selected representative RGBW responses and deformation maps colored by displacement increment.

Fig. 4: Six-axis load calibration and reconstruction. (a) Calibration setup with a reference F/T sensor. (b) 20-channel RGBW preprocessing and MLP-based reconstruction. (c) Training loss and independent test results, including per-axis $R^{2}$ , absolute error distributions, and predicted–reference comparisons.

Fig. 5: Real-time six-axis force and moment reconstruction under combined loading. Reference measurements are shown as red dashed lines and smoothed predictions as blue solid lines, with the mean absolute error (MAE) annotated for each axis. Time is in seconds.

Fig. 6: Normal force responses under weight loading and rapid tapping. (a) Concatenated records of loading with 20 g, 10 g, and 50 g weights; (b) rapid finger tapping. Time is in seconds.

Fig. 7: Interactive manipulation of a simulated Franka robotic arm using LumiTac sensor inputs. (a) Handheld sensor input; (b) planar translation, downward grasping, and rotational alignment operations; (c) corresponding $F_{x}$ , $F_{y}$ , $F_{z}$ , and $M_{z}$ control inputs. Time is in seconds.

Fig. 8: Human–robot interaction experiments using LumiTac. (a) Integration of LumiTac into the robotic gripper and manual load application; (b) admittance-based control framework; (c) translational motion and gripper aperture responses; (d) rotational motion responses. Time is in seconds.

Tactile Sensing

Tactile sensing measures contact forces, deformation, texture, and motion to help robots perceive and act through physical interaction.

See all 146 papers

Aug 31, 2026

SpectraTac: A Compact Camera-Free Optical Tactile Sensor with Distributed Color Sensing

Tactile sensing is essential for physical interaction in robotics and human--machine systems. However, combining rich tactile information with compact hardware, low cost, and low computational overhead remains challenging. This work presents SpectraTac, a compact, camera-free optical tactile sensor that combines active red--green--blue (RGB) illumination with spatially distributed color sensing. Contact deforms a compliant transparent elastomer and modulates its internal light field, producing spatially differentiated changes in color and intensity. Three distributed color sensors capture these responses as low-dimensional spatio-spectral features, avoiding cameras, imaging optics, and high-dimensional image processing. The device measures 19.2 mm in diameter and 4 mm in height, with a material cost below USD~5. A data-driven decoding framework simultaneously estimates three-dimensional (3D) force and the contact region from the optical measurements. For 3D force prediction, the sensor achieved mean absolute errors (MAEs) of 0.161, 0.164, and 0.429 N along the x-, y-, and z-axes, respectively. The nine-region contact-classification accuracy was 99.9%. We further evaluated real-time 3D force tracking and contact-region-based human--machine interaction through an interactive control task. These results indicate that distributed color-resolved optical sensing offers a compact, low-cost alternative to camera-based tactile sensing for robotics, wearable sensing, and interactive systems.

Hao Wu, Haotian Guo, Yu Feng +3

Department of Mechanical Engineering, National University of Singapore, Singapore · Department of Mechanical Engineering, Zhejiang University, Hangzhou, China

Tactile Sensing Robotic Tactile Sensing

Ask PDF

Nov 2, 2026

PanoTac: A Dense Three-Axis Tactile Sensor for Robust Touch-Centric Manipulation

Contact-rich manipulation benefits from tactile feedback that captures both detailed contact geometry and interaction forces. When scooping grains, contact patterns help infer spoon orientation, while resistance transmitted through the grasp guides motion. To provide these complementary sensing capabilities, we present PanoTac, a low-cost, compliant tactile fingertip designed for dense three-axis tactile sensing. Each fingertip provides a

12 \times 32 \times 3

tactile representation, with one normal (

z

) and two shear (

x

y

) response components at each taxel. The sensor operates at 60 Hz and costs less than $50 per fingertip. Sensor characterization quantifies response repeatability and consistency across sensor instances and contact locations. We integrate PanoTac into a robotic system with a visuo-tactile pipeline. Across test-tube preparation, grain scooping, and salt pouring, the resulting policies outperform vision-only baselines by an average of 46.7 percentage points. Ablations show that removing either spatial contact information or shear feedback reduces overall performance across the task suite. Our project page is available at https://panotac.github.io.

Binglin, Wang, Xiaole +7

Andre · Alex · City University of New York +2

Tactile Sensing for Robotic Manipulation Tactile Sensing Contact-Rich Robotic Manipulation

Ask PDF

Apr 30, 2026

FlexiTac: A Low-Cost, Open-Source, Scalable Tactile Sensing Solution for Robotic Systems

We present FlexiTac, a low-cost, open-source, and scalable piezoresistive tactile sensing solution designed for robotic end-effectors. FlexiTac is a practical "plug-in" module consisting of (i) thin, flexible tactile sensor pads that provide dense tactile signals and (ii) a compact multi-channel readout board that streams synchronized measurements for real-time control and large-scale data collection. FlexiTac pads adopt a sealed three-layer laminate stack (FPC-Velostat-FPC) with electrode patterns directly integrated into flexible printed circuits, substantially improving fabrication throughput and repeatability while maintaining mechanical compliance for deployment on both rigid and soft grippers. The readout electronics use widely available, low-cost components and stream tactile signals to a host computer at 100 Hz via serial communication. Across multiple configurations, including fingertip pads and larger tactile mats, FlexiTac can be mounted on diverse platforms without major mechanical redesign. We further show that FlexiTac supports modern tactile learning pipelines, including 3D visuo-tactile fusion for contact-aware decision making, cross-embodiment skill transfer, and real-to-sim-to-real fine-tuning with GPU-parallel tactile simulation. Our project page is available at https://flexitac.github.io/.

Binghao Huang, Yunzhu Li

Columbia University

Tactile Sensing Tactile Sensing for Robotic Manipulation Robotic Tactile Sensing

Ask PDF

Abstract

Tactile Sensing

Explore similar work

SpectraTac: A Compact Camera-Free Optical Tactile Sensor with Distributed Color Sensing

PanoTac: A Dense Three-Axis Tactile Sensor for Robust Touch-Centric Manipulation

FlexiTac: A Low-Cost, Open-Source, Scalable Tactile Sensing Solution for Robotic Systems