Mixtac: A Novel Bio-Inspired Hybrid Tactile Sensor with Synergistic Event-Frame Perception

Bin He; Junkai Xu; Na Ningguta; Peter B. Shull; Shuo Jiang; Yihang Li; Yijin Chen

arxiv: 2605.18550 · v1 · pith:PWOQ4IARnew · submitted 2026-05-18 · 📡 eess.IV

Mixtac: A Novel Bio-Inspired Hybrid Tactile Sensor with Synergistic Event-Frame Perception

Yihang Li , Yijin Chen , Junkai Xu , Na Ningguta , Peter B. Shull , Shuo Jiang , Bin He This is my paper

Pith reviewed 2026-05-20 08:05 UTC · model grok-4.3

classification 📡 eess.IV

keywords tactile sensorevent-based sensinghybrid sensorforce estimationrobotic manipulationbio-inspiredneural network fusion

0 comments

The pith

Mixtac hybrid sensor estimates normal forces with 0.04 N error by fusing high-speed events and stable frames.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes a bio-inspired hybrid tactile sensor that combines event-based and frame-based data to overcome the low sampling rates of vision sensors and the drift problems of event sensors. Events handle rapid force changes while frames provide stable references over time. A network called FGER-Net fuses the streams by using frames to correct drift in training and then guiding fast predictions at runtime. This yields accurate normal force estimates at rates up to 500 Hz, supporting more capable robotic object handling.

Core claim

The novel hybrid event frame tactile sensor Mixtac leverages events for high frequency force tracking and frames for long term accuracy. The Frame Guided Event Recurrent Network fuses the two data streams, with frames correcting event drift during training and guiding high frequency predictions during inference, resulting in a mean absolute error of 0.04 N for normal force estimation.

What carries the argument

Frame Guided Event Recurrent Network (FGER-Net), which fuses event and frame streams so frames correct drift in training and support high-speed output at inference time.

If this is right

Bridges the sampling rate gap from 0 to 500 Hz found in current vision-based tactile sensors.
Supports normal force estimation accurate enough for human-level robotic manipulation.
Demonstrates a practical prototype that emulates synergistic mechanoreceptor function.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same fusion idea could be tested on other robotic sensors that combine fast and slow data streams.
Longer field trials in varied grasping tasks would show whether accuracy holds outside controlled experiments.

Load-bearing premise

The network can reliably fix drifting event signals using frame data in training while keeping fast performance and avoiding new errors when running live.

What would settle it

Run the sensor for long static contacts using only event input and check whether force error stays below 0.04 N or begins to drift.

Figures

Figures reproduced from arXiv: 2605.18550 by Bin He, Junkai Xu, Na Ningguta, Peter B. Shull, Shuo Jiang, Yihang Li, Yijin Chen.

**Figure 2.** Figure 2: The architecture of the proposed FGER-Net is presented, alongside its conceptual underpinnings. The left portion illustrates the biological [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Experimental setup and results for high-frequency vibration sens [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

**Figure 4.** Figure 4: Time-binned violin plot of rolling MAE for normal-force estimation [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

**Figure 5.** Figure 5: Results of different input signals. The predicted normal force is compared against the ground truth (blue) for three input modalities. The [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗

**Figure 6.** Figure 6: Performance comparison for normal force estimation between [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗

**Figure 7.** Figure 7: Results for tracking a hybrid force profile. A 20 g cylindrical [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗

**Figure 8.** Figure 8: Closed loop slip control experiment. (a) The test object, marked [PITH_FULL_IMAGE:figures/full_fig_p008_8.png] view at source ↗

read the original abstract

Vision based and event based tactile sensors are important in robotic manipulation research. However, they suffer from a fundamental tradeoff: vision based sensors have low sampling rates, while event based sensors are prone to drift during long term static force estimation. To solve this challenge and achieve human level tactile perception, the novel hybrid event frame tactile sensor (Mixtac) is proposed in this paper by emulating the synergistic function of biological mechanoreceptors, which achieves normal force estimation. The prototype leverages events for high frequency force tracking and frames for long term accuracy. The Frame Guided Event Recurrent Network (FGER-Net) was proposed to fuse the two data streams. Frames were used by the net to correct event drift during training and guide high frequency predictions during inference. Experiments demonstrated an MAE of 0.04 N. This paper could bridge the sampling rate gap from 0 to 500 Hz in current vision based tactile sensors and pave the way for human level robotic manipulation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Mixtac proposes a hybrid event-frame tactile sensor with FGER-Net fusion to fix drift while keeping high speed, but the reported 0.04 N MAE lacks enough experimental detail to confirm the frequency claim holds up.

read the letter

The core idea is a practical hybrid tactile sensor called Mixtac that pairs event data for fast force changes with frame data for drift-free static readings, using a recurrent network to blend them. They built a prototype and report normal force estimation down to 0.04 N MAE while targeting the 0-500 Hz range that single-modality sensors miss. That addresses a real limitation in robotic manipulation hardware. The bio-inspired framing around mechanoreceptors is straightforward and helps motivate why the two streams should complement each other. The FGER-Net design, where frames correct events in training and guide predictions at inference, is a specific architectural choice that goes beyond generic fusion. They actually built and tested hardware rather than staying in simulation, which gives the work some grounding. The abstract and results section show they measured performance across different force profiles, which is better than many sensor papers that skip real-world checks. The stress-test worry about the recurrent guidance possibly low-pass filtering the event stream at inference time is worth taking seriously. If the hidden-state modulation or attention pulls too much from the slower frames, the output bandwidth could drop even when frames are absent, which would undercut the main selling point. The paper does not appear to include an explicit ablation that isolates the event-only inference bandwidth or compares against a pure event baseline with the same drift correction applied differently. Without those controls or error bars on the high-frequency trials, it is hard to know whether the low MAE comes from true hybrid synergy or from the network averaging things out. The experimental description stays high-level on setup, sensor calibration, and data collection protocol, so a referee would need to press for those. This paper is aimed at roboticists working on tactile sensing for dexterous manipulation who already know the event-versus-frame tradeoff. Someone building their own sensor or fusion pipeline could borrow the Mixtac layout and the FGER-Net structure as a starting point. It is not ready to cite as settled fact yet, but the hardware prototype and the concrete network proposal make it worth a full review rather than a desk reject. I would send it to referees who know both event cameras and force sensing hardware so they can check whether the fusion really preserves the claimed temporal resolution.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes Mixtac, a bio-inspired hybrid tactile sensor that integrates event-based and frame-based vision modalities to estimate normal forces. Events are used for high-frequency force tracking while frames address long-term drift; the Frame Guided Event Recurrent Network (FGER-Net) fuses the streams by employing frames to correct event drift during training and to guide high-frequency predictions during inference. Experiments are reported to achieve a mean absolute error (MAE) of 0.04 N and to bridge the 0–500 Hz sampling-rate gap present in existing vision-based tactile sensors.

Significance. If the central claims hold, the work would represent a meaningful advance in robotic tactile sensing by combining the complementary strengths of event and frame data in a single sensor. The bio-inspired design and the proposed recurrent fusion architecture are conceptually attractive and could support more dexterous manipulation if high temporal bandwidth is demonstrably preserved at inference time.

major comments (2)

[FGER-Net fusion mechanism] FGER-Net description (Section 3 / 4): the mechanism by which frame data guides high-frequency event predictions at inference time is not specified in sufficient detail. If recurrent guidance (hidden-state modulation, attention, or similar) incorporates any averaging or low-pass component derived from the lower-rate frames, the hybrid output will exhibit reduced bandwidth even when frames are absent, directly undermining the claim that the sensor bridges the full 0–500 Hz range while achieving the reported 0.04 N MAE.
[Experimental validation] Experimental results (Section 5): the abstract and results section report an MAE of 0.04 N yet provide no information on sensor calibration, force-application protocol, baseline methods, number of trials, error bars, data-exclusion criteria, or cross-validation procedure. These omissions make it impossible to assess whether the quantitative claim is supported by the data.

minor comments (2)

[Abstract] The abstract would be strengthened by a single sentence describing the physical prototype (e.g., camera model, elastomer thickness, or event-camera resolution) to give readers immediate context for the hybrid hardware.
[Methods] Notation for event and frame streams is introduced without an explicit diagram or equation defining the input tensors; a small schematic would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and will revise the manuscript to improve clarity and completeness where needed.

read point-by-point responses

Referee: [FGER-Net fusion mechanism] FGER-Net description (Section 3 / 4): the mechanism by which frame data guides high-frequency event predictions at inference time is not specified in sufficient detail. If recurrent guidance (hidden-state modulation, attention, or similar) incorporates any averaging or low-pass component derived from the lower-rate frames, the hybrid output will exhibit reduced bandwidth even when frames are absent, directly undermining the claim that the sensor bridges the full 0–500 Hz range while achieving the reported 0.04 N MAE.

Authors: We appreciate the referee's concern about preserving high temporal bandwidth. In FGER-Net, frame data provides sparse, periodic corrections to the recurrent hidden state to counteract event drift at inference time; these corrections are implemented as additive updates without averaging, low-pass filtering, or downsampling of the event-driven pathway. The high-frequency event branch continues to process asynchronous events independently at their native rate. We will revise Section 4 to include explicit equations for the guidance step, a detailed diagram of the inference flow, and an analysis confirming that output bandwidth remains event-limited (up to 500 Hz) when frames are absent. revision: yes
Referee: [Experimental validation] Experimental results (Section 5): the abstract and results section report an MAE of 0.04 N yet provide no information on sensor calibration, force-application protocol, baseline methods, number of trials, error bars, data-exclusion criteria, or cross-validation procedure. These omissions make it impossible to assess whether the quantitative claim is supported by the data.

Authors: We agree that the experimental details are currently insufficient for full reproducibility and evaluation. The revised manuscript will expand Section 5 with a dedicated experimental protocol subsection describing: (i) calibration against a reference force gauge, (ii) the robotic indentation protocol with controlled force profiles and speeds, (iii) comparisons to event-only and frame-only baselines, (iv) 50 trials per condition with standard error bars in all plots, (v) data exclusion criteria (outliers >3σ), and (vi) the 5-fold cross-validation procedure yielding the reported 0.04 N MAE. These additions will directly support the quantitative claims. revision: yes

Circularity Check

0 steps flagged

No circularity: claims rest on experimental results

full rationale

The paper introduces a hybrid event-frame tactile sensor and the FGER-Net fusion architecture, then reports empirical performance (MAE of 0.04 N) from prototype experiments. No equations, parameter fits, or derivations are shown that reduce to their own inputs by construction. The description of frames correcting drift in training while guiding inference is a conventional train/test distinction for recurrent networks and does not create self-definition or fitted-input-as-prediction circularity. No self-citations are invoked as load-bearing uniqueness theorems. The central result is therefore externally falsifiable via replication of the hardware and network training, placing the work in the self-contained category.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Based on abstract only, no explicit free parameters, axioms, or invented entities beyond the sensor prototype and network are detailed; the bio-emulation is presented as inspiration rather than a formal axiom.

invented entities (1)

FGER-Net no independent evidence
purpose: Fuse event and frame data streams for drift correction and high-frequency prediction
Introduced as a novel network in the abstract to enable the hybrid sensing; no independent evidence provided.

pith-pipeline@v0.9.0 · 5722 in / 1119 out tokens · 51387 ms · 2026-05-20T08:05:04.703186+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

The Frame Guided Event Recurrent Network (FGER-Net) was proposed to fuse the two data streams. Frames were used by the net to correct event drift during training and guide high frequency predictions during inference.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages

[1]

Enhancing generalizable 6d pose tracking of an in-hand object with tactile sensing,

Y . Liu, X. Xu, W. Chen, H. Yuan, H. Wang, J. Xu, R. Chen, and L. Yi, “Enhancing generalizable 6d pose tracking of an in-hand object with tactile sensing,”IEEE Robotics and Automation Letters, vol. 9, no. 2, pp. 1106–1113, 2023

work page 2023
[2]

Tactile dexterity: Manipulation primitives with tactile feedback,

F. R. Hogan, J. Ballester, S. Dong, and A. Rodriguez, “Tactile dexterity: Manipulation primitives with tactile feedback,” in2020 IEEE interna- tional conference on robotics and automation (ICRA), pp. 8863–8869. IEEE, 2020

work page 2020
[3]

Tactile sensing for dexter- ous in-hand manipulation in robotics—a review,

H. Yousef, M. Boukallel, and K. Althoefer, “Tactile sensing for dexter- ous in-hand manipulation in robotics—a review,”Sensors and Actuators A: physical, vol. 167, no. 2, pp. 171–187, 2011

work page 2011
[4]

Hardware Technology of Vision-Based Tactile Sensor: A Review,

S. Zhang, Z. Chen, Y . Gao, W. Wan, J. Shan, H. Xue, F. Sun, Y . Yang, and B. Fang, “Hardware Technology of Vision-Based Tactile Sensor: A Review,”IEEE Sensors Journal, vol. 22, DOI 10.1109/JSEN.2022.3210210, no. 22, pp. 21 410–21 427, Nov. 2022

work page doi:10.1109/jsen.2022.3210210 2022
[5]

Robotic tactile perception of object properties: A review,

S. Luo, J. Bimbo, R. Dahiya, and H. Liu, “Robotic tactile perception of object properties: A review,”Mechatronics, vol. 48, pp. 54–67, 2017

work page 2017
[6]

Design, modeling, and validation of a soft magnetic 3-d force sensor,

A. Dwivedi, A. Ramakrishnan, A. Reddy, K. Patel, S. Ozel, and C. D. Onal, “Design, modeling, and validation of a soft magnetic 3-d force sensor,”IEEE Sensors Journal, vol. 18, no. 9, pp. 3852–3863, 2018

work page 2018
[7]

A review of tactile sensing technologies with applications in biomedical engineering,

M. I. Tiwana, S. J. Redmond, and N. H. Lovell, “A review of tactile sensing technologies with applications in biomedical engineering,”Sen- sors and Actuators A: physical, vol. 179, pp. 17–31, 2012

work page 2012
[8]

Human-inspired robotic grasp control with tactile sensing,

J. M. Romano, K. Hsiao, G. Niemeyer, S. Chitta, and K. J. Kuchen- becker, “Human-inspired robotic grasp control with tactile sensing,” IEEE Transactions on Robotics, vol. 27, no. 6, pp. 1067–1079, 2011

work page 2011
[9]

A novel dynamic-vision-based approach for tactile sensing applications,

F. B. Naeini, A. M. AlAli, R. Al-Husari, A. Rigi, M. K. Al-Sharman, D. Makris, and Y . Zweiri, “A novel dynamic-vision-based approach for tactile sensing applications,”IEEE Transactions on Instrumentation and Measurement, vol. 69, no. 5, pp. 1881–1893, 2019

work page 2019
[10]

Portable high-speed optical gaze con- troller with vision chip,

L. Miyashita and M. Ishikawa, “Portable high-speed optical gaze con- troller with vision chip,”Journal of Robotics and Mechatronics, vol. 34, no. 5, pp. 1133–1140, 2022

work page 2022
[11]

Evetac: An event-based optical tactile sensor for robotic manipulation,

N. Funk, E. Helmut, G. Chalvatzaki, R. Calandra, and J. Peters, “Evetac: An event-based optical tactile sensor for robotic manipulation,”IEEE Transactions on Robotics, 2024

work page 2024
[12]

E-BTS: Event-Based Tactile Sensor for Haptic Teleoperation in Augmented Reality,

D. Mukashev, S. Seitzhan, J. Chumakov, S. Khajikhanov, M. Yergibay, N. Zhaniyar, R. Chibar, A. Mazhitov, M. Rubagotti, and Z. Kap- passov, “E-BTS: Event-Based Tactile Sensor for Haptic Teleoperation in Augmented Reality,”IEEE Transactions on Robotics, vol. 41, DOI 10.1109/TRO.2024.3502215, pp. 450–463, 2025

work page doi:10.1109/tro.2024.3502215 2024
[13]

Gelevent—a novel high-speed tactile sensor with event cam- era,

D. Yin, S. Lu, J. Yang, Y . Zhang, Z. Dai, D. Nan, B. Cai, S. He, and X. Chen, “Gelevent—a novel high-speed tactile sensor with event cam- era,”IEEE Transactions on Instrumentation and Measurement, 2025

work page 2025
[14]

Microsaccade-inspired event camera for robotics,

B. He, Z. Wang, Y . Zhou, J. Chen, C. D. Singh, H. Li, Y . Gao, S. Shen, K. Wang, Y . Caoet al., “Microsaccade-inspired event camera for robotics,”Science Robotics, vol. 9, no. 90, p. eadj8124, 2024

work page 2024
[15]

Neuromorphic vision-based tactile sensor for robotic grasp,

F. Baghaei Naeini, “Neuromorphic vision-based tactile sensor for robotic grasp,” Ph.D. dissertation, Kingston University, 2020

work page 2020
[16]

Wasti,Spatiotemporal Alignment of Event Stream with Images and Active Illumination

A. Wasti,Spatiotemporal Alignment of Event Stream with Images and Active Illumination. Rochester Institute of Technology, 2024

work page 2024
[17]

Retain, blend, and exchange: A quality-aware spatial- stereo fusion approach for event stream recognition,

L. Chen, D. Li, X. Wang, P. Shao, W. Zhang, Y . Wang, Y . Tian, and J. Tang, “Retain, blend, and exchange: A quality-aware spatial- stereo fusion approach for event stream recognition,”arXiv preprint arXiv:2406.18845, 2024

work page arXiv 2024
[18]

The tactip family: Soft optical tactile sensors with 3d-printed biomimetic morphologies,

B. Ward-Cherrier, N. Pestell, L. Cramphorn, B. Winstone, M. E. Gian- naccini, J. Rossiter, and N. F. Lepora, “The tactip family: Soft optical tactile sensors with 3d-printed biomimetic morphologies,”Soft robotics, vol. 5, no. 2, pp. 216–227, 2018

work page 2018
[19]

In- hand object localization using a novel high-resolution visuotactile sensor,

S. Cui, R. Wang, J. Hu, J. Wei, S. Wang, and Z. Lou, “In- hand object localization using a novel high-resolution visuotactile sensor,”IEEE Transactions on Industrial Electronics, vol. 69, DOI 10.1109/TIE.2021.3090697, no. 6, pp. 6015–6025, 2022. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

work page doi:10.1109/tie.2021.3090697 2021
[20]

Development of an optical sensor capable of measuring distance, tilt, and contact force,

T. Nozaki and H. I. Krebs, “Development of an optical sensor capable of measuring distance, tilt, and contact force,”IEEE Transactions on Industrial Electronics, vol. 69, DOI 10.1109/TIE.2021.3084168, no. 5, pp. 4938–4945, 2022

work page doi:10.1109/tie.2021.3084168 2021
[21]

Gel- stereo 2.0: An improved gelstereo sensor with multimedium refractive stereo calibration,

C. Zhang, S. Cui, S. Wang, J. Hu, Y . Cai, R. Wang, and Y . Wang, “Gel- stereo 2.0: An improved gelstereo sensor with multimedium refractive stereo calibration,”IEEE Transactions on Industrial Electronics, vol. 71, DOI 10.1109/TIE.2023.3312418, no. 7, pp. 7452–7462, 2024

work page doi:10.1109/tie.2023.3312418 2023
[22]

A highly sensitive, compact, and low-cost soft triaxial optical force (stof) sensor for robotic grasping and human–machine interaction,

L. Mo, Y . Li, S. Fu, J. Xia, Y .-F. Zhang, Y . Zhong, X. Chen, and J. G. Chase, “A highly sensitive, compact, and low-cost soft triaxial optical force (stof) sensor for robotic grasping and human–machine interaction,”IEEE Transactions on Industrial Electronics, vol. 73, DOI 10.1109/TIE.2025.3603085, no. 1, pp. 1469–1480, 2026

work page doi:10.1109/tie.2025.3603085 2025
[23]

9dtact: A compact vision-based tactile sensor for accurate 3d shape reconstruction and generalizable 6d force estimation,

C. Lin, H. Zhang, J. Xu, L. Wu, and H. Xu, “9dtact: A compact vision-based tactile sensor for accurate 3d shape reconstruction and generalizable 6d force estimation,”IEEE Robotics and Automation Letters, vol. 9, no. 2, pp. 923–930, 2023

work page 2023
[24]

Front and back illuminated dynamic and active pixel vision sensors comparison,

G. Taverni, D. P. Moeys, C. Li, C. Cavaco, V . Motsnyi, D. S. S. Bello, and T. Delbruck, “Front and back illuminated dynamic and active pixel vision sensors comparison,”IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 65, no. 5, pp. 677–681, 2018

work page 2018
[25]

A soft thumb-sized vision- based sensor with accurate all-round force perception,

H. Sun, K. J. Kuchenbecker, and G. Martius, “A soft thumb-sized vision- based sensor with accurate all-round force perception,”Nature Machine Intelligence, vol. 4, no. 2, pp. 135–145, 2022

work page 2022
[26]

Zhang, W

D. Zhang, W. Fan, J. Lin, H. Li, Q. Cong, W. Liu, N. F. Lepora, and S. Luo, “Design and benchmarking of a multi-modality sensor Yihang Lireceived the B.E. degree in robotics engineering from Harbin Engineering University, in 2025, China. He is currently working toward the Ph.D. degree at the Shanghai Research Institute for Intelligent Autonomous Systems, ...

work page 2025
[27]

candidate at Shanghai Jiao Tong Uni- versity, China

From 2014 to 2020, he studied as a Ph.D. candidate at Shanghai Jiao Tong Uni- versity, China. He is with IMU Master Technol- ogy Co., Ltd, Shanghai 200240, China, focusing on embedded system development, hardware- software co-debugging, and performance opti- mization. for robotic manipulation with gan-based cross-modality interpretation,” IEEE Transaction...

work page 2014

[1] [1]

Enhancing generalizable 6d pose tracking of an in-hand object with tactile sensing,

Y . Liu, X. Xu, W. Chen, H. Yuan, H. Wang, J. Xu, R. Chen, and L. Yi, “Enhancing generalizable 6d pose tracking of an in-hand object with tactile sensing,”IEEE Robotics and Automation Letters, vol. 9, no. 2, pp. 1106–1113, 2023

work page 2023

[2] [2]

Tactile dexterity: Manipulation primitives with tactile feedback,

F. R. Hogan, J. Ballester, S. Dong, and A. Rodriguez, “Tactile dexterity: Manipulation primitives with tactile feedback,” in2020 IEEE interna- tional conference on robotics and automation (ICRA), pp. 8863–8869. IEEE, 2020

work page 2020

[3] [3]

Tactile sensing for dexter- ous in-hand manipulation in robotics—a review,

H. Yousef, M. Boukallel, and K. Althoefer, “Tactile sensing for dexter- ous in-hand manipulation in robotics—a review,”Sensors and Actuators A: physical, vol. 167, no. 2, pp. 171–187, 2011

work page 2011

[4] [4]

Hardware Technology of Vision-Based Tactile Sensor: A Review,

S. Zhang, Z. Chen, Y . Gao, W. Wan, J. Shan, H. Xue, F. Sun, Y . Yang, and B. Fang, “Hardware Technology of Vision-Based Tactile Sensor: A Review,”IEEE Sensors Journal, vol. 22, DOI 10.1109/JSEN.2022.3210210, no. 22, pp. 21 410–21 427, Nov. 2022

work page doi:10.1109/jsen.2022.3210210 2022

[5] [5]

Robotic tactile perception of object properties: A review,

S. Luo, J. Bimbo, R. Dahiya, and H. Liu, “Robotic tactile perception of object properties: A review,”Mechatronics, vol. 48, pp. 54–67, 2017

work page 2017

[6] [6]

Design, modeling, and validation of a soft magnetic 3-d force sensor,

A. Dwivedi, A. Ramakrishnan, A. Reddy, K. Patel, S. Ozel, and C. D. Onal, “Design, modeling, and validation of a soft magnetic 3-d force sensor,”IEEE Sensors Journal, vol. 18, no. 9, pp. 3852–3863, 2018

work page 2018

[7] [7]

A review of tactile sensing technologies with applications in biomedical engineering,

M. I. Tiwana, S. J. Redmond, and N. H. Lovell, “A review of tactile sensing technologies with applications in biomedical engineering,”Sen- sors and Actuators A: physical, vol. 179, pp. 17–31, 2012

work page 2012

[8] [8]

Human-inspired robotic grasp control with tactile sensing,

J. M. Romano, K. Hsiao, G. Niemeyer, S. Chitta, and K. J. Kuchen- becker, “Human-inspired robotic grasp control with tactile sensing,” IEEE Transactions on Robotics, vol. 27, no. 6, pp. 1067–1079, 2011

work page 2011

[9] [9]

A novel dynamic-vision-based approach for tactile sensing applications,

F. B. Naeini, A. M. AlAli, R. Al-Husari, A. Rigi, M. K. Al-Sharman, D. Makris, and Y . Zweiri, “A novel dynamic-vision-based approach for tactile sensing applications,”IEEE Transactions on Instrumentation and Measurement, vol. 69, no. 5, pp. 1881–1893, 2019

work page 2019

[10] [10]

Portable high-speed optical gaze con- troller with vision chip,

L. Miyashita and M. Ishikawa, “Portable high-speed optical gaze con- troller with vision chip,”Journal of Robotics and Mechatronics, vol. 34, no. 5, pp. 1133–1140, 2022

work page 2022

[11] [11]

Evetac: An event-based optical tactile sensor for robotic manipulation,

N. Funk, E. Helmut, G. Chalvatzaki, R. Calandra, and J. Peters, “Evetac: An event-based optical tactile sensor for robotic manipulation,”IEEE Transactions on Robotics, 2024

work page 2024

[12] [12]

E-BTS: Event-Based Tactile Sensor for Haptic Teleoperation in Augmented Reality,

D. Mukashev, S. Seitzhan, J. Chumakov, S. Khajikhanov, M. Yergibay, N. Zhaniyar, R. Chibar, A. Mazhitov, M. Rubagotti, and Z. Kap- passov, “E-BTS: Event-Based Tactile Sensor for Haptic Teleoperation in Augmented Reality,”IEEE Transactions on Robotics, vol. 41, DOI 10.1109/TRO.2024.3502215, pp. 450–463, 2025

work page doi:10.1109/tro.2024.3502215 2024

[13] [13]

Gelevent—a novel high-speed tactile sensor with event cam- era,

D. Yin, S. Lu, J. Yang, Y . Zhang, Z. Dai, D. Nan, B. Cai, S. He, and X. Chen, “Gelevent—a novel high-speed tactile sensor with event cam- era,”IEEE Transactions on Instrumentation and Measurement, 2025

work page 2025

[14] [14]

Microsaccade-inspired event camera for robotics,

B. He, Z. Wang, Y . Zhou, J. Chen, C. D. Singh, H. Li, Y . Gao, S. Shen, K. Wang, Y . Caoet al., “Microsaccade-inspired event camera for robotics,”Science Robotics, vol. 9, no. 90, p. eadj8124, 2024

work page 2024

[15] [15]

Neuromorphic vision-based tactile sensor for robotic grasp,

F. Baghaei Naeini, “Neuromorphic vision-based tactile sensor for robotic grasp,” Ph.D. dissertation, Kingston University, 2020

work page 2020

[16] [16]

Wasti,Spatiotemporal Alignment of Event Stream with Images and Active Illumination

A. Wasti,Spatiotemporal Alignment of Event Stream with Images and Active Illumination. Rochester Institute of Technology, 2024

work page 2024

[17] [17]

Retain, blend, and exchange: A quality-aware spatial- stereo fusion approach for event stream recognition,

L. Chen, D. Li, X. Wang, P. Shao, W. Zhang, Y . Wang, Y . Tian, and J. Tang, “Retain, blend, and exchange: A quality-aware spatial- stereo fusion approach for event stream recognition,”arXiv preprint arXiv:2406.18845, 2024

work page arXiv 2024

[18] [18]

The tactip family: Soft optical tactile sensors with 3d-printed biomimetic morphologies,

B. Ward-Cherrier, N. Pestell, L. Cramphorn, B. Winstone, M. E. Gian- naccini, J. Rossiter, and N. F. Lepora, “The tactip family: Soft optical tactile sensors with 3d-printed biomimetic morphologies,”Soft robotics, vol. 5, no. 2, pp. 216–227, 2018

work page 2018

[19] [19]

In- hand object localization using a novel high-resolution visuotactile sensor,

S. Cui, R. Wang, J. Hu, J. Wei, S. Wang, and Z. Lou, “In- hand object localization using a novel high-resolution visuotactile sensor,”IEEE Transactions on Industrial Electronics, vol. 69, DOI 10.1109/TIE.2021.3090697, no. 6, pp. 6015–6025, 2022. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS

work page doi:10.1109/tie.2021.3090697 2021

[20] [20]

Development of an optical sensor capable of measuring distance, tilt, and contact force,

T. Nozaki and H. I. Krebs, “Development of an optical sensor capable of measuring distance, tilt, and contact force,”IEEE Transactions on Industrial Electronics, vol. 69, DOI 10.1109/TIE.2021.3084168, no. 5, pp. 4938–4945, 2022

work page doi:10.1109/tie.2021.3084168 2021

[21] [21]

Gel- stereo 2.0: An improved gelstereo sensor with multimedium refractive stereo calibration,

C. Zhang, S. Cui, S. Wang, J. Hu, Y . Cai, R. Wang, and Y . Wang, “Gel- stereo 2.0: An improved gelstereo sensor with multimedium refractive stereo calibration,”IEEE Transactions on Industrial Electronics, vol. 71, DOI 10.1109/TIE.2023.3312418, no. 7, pp. 7452–7462, 2024

work page doi:10.1109/tie.2023.3312418 2023

[22] [22]

A highly sensitive, compact, and low-cost soft triaxial optical force (stof) sensor for robotic grasping and human–machine interaction,

L. Mo, Y . Li, S. Fu, J. Xia, Y .-F. Zhang, Y . Zhong, X. Chen, and J. G. Chase, “A highly sensitive, compact, and low-cost soft triaxial optical force (stof) sensor for robotic grasping and human–machine interaction,”IEEE Transactions on Industrial Electronics, vol. 73, DOI 10.1109/TIE.2025.3603085, no. 1, pp. 1469–1480, 2026

work page doi:10.1109/tie.2025.3603085 2025

[23] [23]

9dtact: A compact vision-based tactile sensor for accurate 3d shape reconstruction and generalizable 6d force estimation,

C. Lin, H. Zhang, J. Xu, L. Wu, and H. Xu, “9dtact: A compact vision-based tactile sensor for accurate 3d shape reconstruction and generalizable 6d force estimation,”IEEE Robotics and Automation Letters, vol. 9, no. 2, pp. 923–930, 2023

work page 2023

[24] [24]

Front and back illuminated dynamic and active pixel vision sensors comparison,

G. Taverni, D. P. Moeys, C. Li, C. Cavaco, V . Motsnyi, D. S. S. Bello, and T. Delbruck, “Front and back illuminated dynamic and active pixel vision sensors comparison,”IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 65, no. 5, pp. 677–681, 2018

work page 2018

[25] [25]

A soft thumb-sized vision- based sensor with accurate all-round force perception,

H. Sun, K. J. Kuchenbecker, and G. Martius, “A soft thumb-sized vision- based sensor with accurate all-round force perception,”Nature Machine Intelligence, vol. 4, no. 2, pp. 135–145, 2022

work page 2022

[26] [26]

Zhang, W

D. Zhang, W. Fan, J. Lin, H. Li, Q. Cong, W. Liu, N. F. Lepora, and S. Luo, “Design and benchmarking of a multi-modality sensor Yihang Lireceived the B.E. degree in robotics engineering from Harbin Engineering University, in 2025, China. He is currently working toward the Ph.D. degree at the Shanghai Research Institute for Intelligent Autonomous Systems, ...

work page 2025

[27] [27]

candidate at Shanghai Jiao Tong Uni- versity, China

From 2014 to 2020, he studied as a Ph.D. candidate at Shanghai Jiao Tong Uni- versity, China. He is with IMU Master Technol- ogy Co., Ltd, Shanghai 200240, China, focusing on embedded system development, hardware- software co-debugging, and performance opti- mization. for robotic manipulation with gan-based cross-modality interpretation,” IEEE Transaction...

work page 2014