Flexible Tactile Sensors for Enhancing Robotic Perception

Huanhuan Ren, Wangyang Li, Yanan Ding, Yuming Feng, Hexin Li, Jianyang Li, Jia Zhang, Haiying Xiao, Jialu Li, Shuai Wang, Cong Huang, Li Jiang, PingAn HuSmart 2025 · DOI

TL;DR

This review explores the transition of robotic perception from rigid, limited sensors to flexible, "iontronic," and multimodal tactile skins. By leveraging biomimetic microstructures and advanced materials, researchers are enabling robots to "feel" textures, temperatures, and microscopic forces with human-like sensitivity, which is critical for the future of autonomous surgery and precision manufacturing.

What Problem Does This Paper Address?

While computer vision has given robots incredible "sight," their sense of touch—tactile perception—has lagged behind. Traditional tactile sensors are often rigid, bulky, and limited to sensing a single parameter (usually just normal pressure). This creates a bottleneck in environments requiring delicate interaction, such as handling a soft fruit without bruising it or performing minimally invasive surgery.

To interact safely and intelligently with the world, robots need "electronic skins" that are flexible, highly sensitive, and capable of multimodal sensing—the ability to detect pressure, shear force, temperature, and even humidity simultaneously. The authors tackle the challenge of how to design these complex systems using new materials and structural optimizations.

Methods

The paper provides a comprehensive review of the five primary mechanisms used in modern flexible tactile sensing:

  1. Resistive: Measures changes in contact resistance. While cost-effective, they often struggle with response speed.
  2. Capacitive: Detects changes in geometry. These are precise but often have a limited sensing range.
  3. Piezoelectric: Generates voltage from deformation. Excellent for detecting vibrations and textures (dynamic sensing) but poor at measuring steady, static pressure.
  4. Triboelectric: Uses contact electrification (the "static cling" effect) to create self-powered sensors.
  5. Ion-electronic (Iontronic): A major focus of the paper. These use ionic gels or liquids to create an Electric Double Layer (EDL). Because the "gap" in these capacitors is at the nanoscale, they offer orders-of-magnitude higher sensitivity than traditional capacitive sensors.

The authors also highlight Microstructural Design, where researchers use nature-inspired shapes—like pyramids, domes, or even "spider-web" patterns—to make the sensors more compressible and responsive.

Datasets

As a review paper, this work does not introduce a new dataset. However, it synthesizes results from hundreds of studies. It specifically highlights the importance of large-scale tactile datasets for training AI, such as the 135,000-frame full-hand tactile dataset developed by Sundaram et al. [191], which is used to train deep convolutional neural networks for object recognition and weight estimation.

Key Insights and Ideas

  • The "Iontronic" Advantage: The shift from solid materials to ionic gels is a game-changer. By using the EDL effect, sensors can achieve sensitivities as high as 3302 kPa⁻¹, allowing robots to detect the weight of a tiny insect or the pulse of a human wrist.
  • Biomimetic Optimization: Researchers are increasingly looking to biology. For example, using the surface texture of Calathea zebrina leaves as a template for sensor microstructures helps achieve high sensitivity across a much wider pressure range.
  • Decoupling Signals: One of the hardest problems in tactile sensing is "crosstalk"—when a sensor can't tell the difference between a change in temperature and a change in pressure. The paper discusses "low-coupling" designs that use unique material layers to isolate these signals.
  • AI as the "Brain" for the Skin: Raw tactile data is incredibly noisy. The authors emphasize that intelligent algorithms (like neural networks) are now essential to translate raw electrical signals into meaningful concepts like "roughness," "stiffness," or "slippage."

Key Results

The review compares performance across different technologies (detailed in Table 1):

  • Sensitivity: Ion-electronic sensors (e.g., PVA@H3PO4) lead the pack with sensitivities over 3300 kPa⁻¹, compared to traditional resistive sensors which often hover around 0.1–5 kPa⁻¹.
  • Response Time: Modern flexible sensors are reaching response times as fast as 9 ms, approaching the speed required for real-time robotic reflex actions.
  • Recognition Accuracy: Using triboelectric "smart fingers" combined with machine learning, researchers have achieved a 96.8% accuracy rate in identifying different material types based solely on touch.

Limitations and Future Work

Despite the progress, several hurdles remain:

  • Durability vs. Flexibility: Flexible materials are prone to wear and tear. Developing "self-healing" conductive polymers is a high-priority research area.
  • Sensor Density: Packing thousands of sensors into a small area (like a robotic fingertip) creates massive amounts of data that are difficult to process in real-time.
  • Power Consumption: While triboelectric sensors are self-powered, integrating them into a full-body "skin" with wireless data transmission remains energy-intensive.
  • Future Direction: The authors point toward Neuromorphic Computing—chips that process tactile data locally on the "skin" (similar to how human nerves process some signals before they reach the brain) to reduce latency and power use.

Summary updated

Abstract

As robotic technology advances towards autonomy and intelligence, tactile sensing systems have become a key technology for overcoming the bottleneck of environmental interaction. Unlike visual perception, which captures macroscopic information, tactile sensing provides precise mechanical feedback for robotic manipulation by real-time analysis of microscopic physical parameters such as contact force, material properties, and surface morphology. This is particularly important in application scenarios that require high precision in force control, such as minimally invasive surgery and precision assembly, where the sensitivity, multimodal perception capabilities, and environmental adaptability of tactile sensors directly determine the operational performance of the robotic system. However, traditional tactile sensors are limited by rigid structures, limited sensitivity, and a single perception modality, making it difficult to meet the demands of complex interactions. In recent years, breakthroughs in flexible electronic materials, biomimetic microstructure design, and multimodal sensing integration technologies have provided innovative opportunities for tactile sensing systems. This review provides a detailed overview of the key technological advancements in tactile sensors and their applications in robotic surfaces, analyzes the major challenges currently faced, and looks ahead to future development directions, particularly the potential in flexible materials and intelligent algorithms. Received: 25 March 2025 | Revised: 26 May 2025 | Accepted: 28 May 2025 Conflicts of Interest PingAn Hu is the Editor-in-Chief for Smart Wearable Technology and was not involved in the editorial review or the decision to publish this article. The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data sharing is not applicable to this article as no new data were created or analyzed in this study. Author Contribution Statement Huiwen Ren: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Wangyang Li: Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Yanan Ding: Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Yuming Feng: Writing – original draft. Hexin Li: Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Jianyang Li: Writing – original draft, Writing – review & editing. Jia Zhang: Writing – original draft, Writing – review & editing, Supervision. Haiying Xiao: Writing – original draft, Writing – review & editing. Jialu Li: Writing – original draft, Writing – review & editing. Shuai Wang: Writing – original draft. Cong Huang: Writing – original draft, Writing – review & editing, Supervision. Li Jiang: Writing – original draft, Writing – review & editing, Supervision. PingAn Hu: Conceptualization, Methodology, Writing – original draft, Writing – review & editing, Supervision, Project administration, Funding acquisition.
BibTeX
@inproceedings{ren2025flexible,
  author = {Huanhuan Ren and Wangyang Li and Yanan Ding and Yuming Feng and Hexin Li and Jianyang Li and Jia Zhang and Haiying Xiao and Jialu Li and Shuai Wang and Cong Huang and Li Jiang and PingAn Hu},
  title = {{Flexible Tactile Sensors for Enhancing Robotic Perception}},
  booktitle = {Smart},
  year = {2025},
  doi = {10.47852/bonviewswt52025770},
}
Figure