Real-Time Fingertip Force Estimation from Lateral Deformation for Precision Manipulation Tasks

This study proposes a wearable fingertip force estimation approach that employs a single three-axis force sensor mounted on the side of the fingernail, enabling measurements under natural contact conditions without obscuring the fingerpad. Fingertip deformation during contact is captured by the sensor, and calibration experiments are performed using a table-mounted force sensor to relate the measured lateral deformation to actual normal and tangential forces. Two modeling approaches—a 6th-order multi-input multi-output transfer function model and an 8th-order state-space model—are identified from the calibration data. The transfer function model achieves high offline accuracy (R2 ≈ 0.9, RMSE ≤ 7%FS), while the state-space model offers smoother, low-latency output suitable for real-time feedback. These results clarify the relationship between fingertip deformation and applied forces, providing a scientific basis for improving skill transfer, training protocols, and haptic simulation in precision manual tasks. NTRODUCTION The human fingertip is densely populated with tactile receptors, and the mechanical information captured by mechanoreceptors plays a crucial role in gripping and manipulating objects. In precision assembly and machining, highly skilled workers rely on fingertip sensations to adjust applied forces and to judge work quality. However, such tactile skills depend heavily on personal experience and subjective judgment, posing challenges for skill transfer and instruction. Advances in tactile sensing and measurement technologies offer new ways to address this issue. sensor to simulate finger-pad pressing under various conditions [5], establish a direct link between lateral forces and actual finger-pad deformation, and employ transfer-function and state- In this work, a single three-axis tactile force sensor is used to investigate fingertip force regulation and deformation perception under natural contact conditions (Fig.1). Calibration experiments in conjunction with a table-mounted tri-axial force Finger-pad deformation during contact events is typically analyzed with 2-D or 3-D computational models [1]. For example, multilayer 3-D finite-element models have been used to examine stress and strain at receptor locations [2]; simplified fingernail-shell models can efficiently simulate surface deformation of the finger pad [3]; and 3-D beam-bundle models have been applied to simulate fingertip interactions during pressing and sliding on a plane [4]. These studies explore, at a theoretical level, the relationship between tissue deformation of the finger pad and the forces exerted. This Research is supported by the New Energy and Industrial Technology Development Organization (NEDO). Jiang Shixuan is with the Graduate School of Science and Engineering, Kindai University, Osaka, Japan 2433330361h@kindai.ac.jp space models to predict both deformation and force [6]. The objective, compared with earlier studies, is to develop a system that relies on just one lateral tactile sensor, accurately measures finger-pad deformation without disturbing natural manipulation, and maintains a high degree of mounting freedom while minimizing interference with the contact surface. The prediction framework captures dynamic characteristics more effectively by adopting both transfer- function and state-space system-identification approaches, enabling model accuracy to be optimized across different contact scenarios. This capability is critical for enhancing the reliability and accuracy of tactile feedback in high-precision manual tasks such as micro-assembly, conservation of cultural artifacts, and surgical procedures. The anticipated applications of this technology will deepen understanding of human tactile mechanisms, provide practical tools for skill assessment and education, and facilitate effective skill. The following sections will describe the hardware of the FLD sensor and the principles behind measuring lateral fingertip deformation and forces. The method for estimating Real-Time Fingertip Force Estimation from Lateral Deformation Atsutoshi Ikeda is with the Graduate School of Science and Engineering, Kindai University, Osaka, Japan, ikeda@emat.kindai.ac.jp https://orcid.org/0000-0002-7380-4048 for Precision Manipulation Task Jiang Shixuan, and Atsutoshi Ikeda, Member, IEEE Fig. 1 Demonstration of the proposed wearable tactile sensor in a writing task 2026 IEEE/SICE International Symposium on System Integration (SII) January 11-14, 2026. Cancún, México 978-1-6654-5784-2/26/$31.00 ©2026 IEEE 300