Advanced Tactile Sensing: Differentiating Softness through Transient Dynamics
The 14th International Conference on Advanced Materials and Devices, BEXCO Busan, Korea (2025)
Myung Hoe Kim, Sangmin Lee, Suyeon Kang, Young Pyo Jeon, Dongpyo Hong, Sang Yoon Park and Young Joon Yoo
[Abstract]
This study introduces a novel tactile sensor designed to enhance the recognition of material softness, a critical capability for advanced robotics and human-machine interfaces. Traditional tactile sensors primarily measure pressure, which is often insufficient for distinguishing subtle
differences in softness. To overcome this limitation, we developed a sensor utilizing an interlocking pyramid microstructure fabricated from a carbon nanoparticle-
polydimethylsiloxane (CNP-PDMS) composite. The key innovation of our sensor lies in its ability to capture not only the final saturated
pressure but also the transient response during material deformation. This provides a new, critical parameter for softness evaluation: the deformation time (t_d), which is the time taken to reach the maximum indentation depth. This dual-parameter approach emulates the human tactile sense more comprehensively. To validate the sensor's performance, we measured its response to four materials with distinct softness levels (Resin, PDMS, Ecoflex, and TPU-foam) under a consistent pressure of 15 kPa. The results, as shown in Fig. 1, demonstrate that each material exhibits a unique combination of saturated output current and deformation time. We then applied a 1-dimensional convolutional neural network (1D-CNN) for classification. When both pressure and deformation time data were used, the model achieved 100% accuracy in distinguishing the materials. In contrast, using only the saturated pressure data—mimicking conventional sensors—resulted in an accuracy of just 73%, with significant confusion between the softer materials. This work confirms that incorporating transient deformation data is crucial for accurate softness recognition. The proposed sensor, with its unique structure and analytical approach, offers a significant advancement in tactile sensing technology.