Self-powered multimodal tactile sensing enabled by hybrid triboelectric and magnetoelastic mechanisms

Tactile sensing technology, an important branch in the field of intelligent robotics, has received extensive attention in recent years. “Object property perception, as a core component of tactile sensing technology, faces severe challenges due to its inherent complexity and diversity, particularly under the constraints of decoupling difficulty and limited precision.” Said the author. The primary goal of tactile sensing technology is to simulate the tactile function of human skin and achieve the perception of the properties of objects. However, the material, roughness, and softness of an object have the characteristics of variety, interaction complexity, and fuzzy standards, and these 3 properties are usually integrated with each other, which greatly increases the difficulty of decoupling and accurate identification. For each of the above 3 properties, there are corresponding sensing mechanisms that can accurately identify them. Obviously, a single sensing mechanism cannot decouple these 3 properties simultaneously, which leads researchers to focus on multimodal sensing. “However, multimodal sensing is not a simple superposition of sensing mechanisms, and it is necessary to consider the optimal integration mode among sensors and the mutual interference problem between different sensing mechanisms.” said study author Shaorong Xie, a professor at Shanghai University.

Triboelectric sensing, based on triboelectrification and electrostatic induction, contains abundant object property information, can recognize materials and roughness, and has advantages like simple structure, strong expandability, and miniaturization, making it a key basis for constructing MMTSDs based on TENGs. However, most current TENG-based multimodal devices rely on capacitive, resistive, piezoelectric, and triboelectric sensors, which are highly susceptible to environmental factors such as temperature and humidity, and the identification of object roughness is also significantly affected by external forces. “Fortunately, the recently discovered magnetoelastic effect in soft material systems has led to the development of soft magnetoelastic generators (MEGs), which are minimally affected by the environment, have high-sensitivity force-sensing capabilities, and possess self-powered properties. Therefore, using magnetoelastic sensing as the force feedback unit in triboelectric-based tactile sensing devices is an ideal strategy.” emphasized the authors.

In this paper, author have created an MMTSD based on triboelectric–magnetoelastic, which has anthropomorphic tactile sensation. This MMTSD effectively interfaces the TENG array with the MEG through silica gel, enabling it to stably perceive the properties of objects within an open environment. To evaluate the efficacy of MMTSD, we used a mechanical claw as the carrier of the MMTSD, carried out grasping tests on 8 material types, 3 softness levels, 4 roughness levels, and 8 objects with intertwined properties and established a comprehensive electrical signal dataset under environmental change. Furthermore, the author constructed a “material–softness–roughness” perception model based on a lightweight convolutional neural network, achieving a recognition accuracy rate of more than 95% for materials, softness, and roughness (material recognition accuracy rate: 99.07%, softness recognition accuracy rate: 100%, roughness recognition accuracy rate: 95.56%, and comprehensive property recognition accuracy rate: 95.83%).

In summary, this paper introduces an MMTSD by integrating triboelectric and magnetoelastic mechanisms. The MMTSD combines self-power, high precision, and multifunctionality. The MMTSD designed in this study provides a more comprehensive tactile perception ability for robots, improves the intelligence and adaptability of robot operations, and promotes the innovation and development of robots in multiple industries.

Authors of the paper include Xiao Lu, Tianhong Wang, Songyi Zhong, Tianqi Cao, Chenghao Zhou, Long Li, Quan Zhang, Shiwei Tian, Tao Jin, Tao Yue, and Shaorong Xie.

This work has been completed with the financial support of the National Key Research and Development Program of China (grant no. 2023YFB4705200), in part by the National Natural Science Foundation of China (grant nos. 62303291, 62373235, 62473244, and 62273222), in part by the Natural Science Foundation of Shanghai (grant no. 23ZR1423700), and in part by the Foundation of Science and Technology Commission of Shanghai Municipality (grant nos. 24511103800 and 24TS1402300).

The paper, “Self-Powered Multimodal Tactile Sensing Enabled by Hybrid Triboelectric and Magnetoelastic Mechanisms” was published in the journal Cyborg and Bionic Systems on Jul 2, 2025, at DOI: 10.34133/cbsystems.0320.