A Two‐Stage Characterization Pipeline and Open‐Source Framework for Reproducible Tactile Sensing

Journal article

Tactile sensing endows modern machines with a compliant, information‐rich sense of touch, enabling safe human–robot interaction and delicate manipulation. While recent material breakthroughs have yielded diverse soft sensor technologies, viscoelastic creep, mechanical hysteresis, and similar complexities render their electromechanical behaviors nonlinear and difficult to isolate. Consequently, the lack of standardized characterization has fragmented the field, limiting cross‐validation, reproducibility, and system‐level integration. Here, we present a comprehensive two‐stage electromechanical characterization pipeline and an open‐source framework that standardizes evaluation of any deformable tactile sensor whose transduction yields a scalar or low‐dimensional electrical output (resistive, capacitive, inductive, magnetic, piezoelectric, or discrete optical). Unlike ad‐hoc testing, our methodology decouples intrinsic material properties from embodied system performance: Stage 1 (Benchtop Characterization) rigorously isolates the sensor's fundamental electromechanical behaviors under idealized boundaries, while Stage 2 (in situ functional evaluation) captures the embodiment‐modulated response of the sensorized body within its specific interaction regime. To ensure adoption, a FAIR‐compliant reproducibility toolkit accompanies the framework: a machine‐readable digital datasheet schema, algorithmic metric definitions, and a multithreaded controller for deterministic data acquisition. By establishing a common language and a baseline for experimental execution, this framework will accelerate the evolution of tactile sensing into a mature and reproducible engineering discipline.

Authors: Lo Preti, M; Buxiu, W; Trunin, P; Nazeer, MS; Beccai, L

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Wiley
• Journal: Advanced Intelligent Systems
• Article number: e70436
• Publication date: 2026-06-01
• Acceptance date: 2026-05-14
• DOI: 10.1002/aisy.70436
• EISSN: 2640-4567
• ISSN: 2640-4567
• Language: English
• Keywords: soft sensing; benchmarking; tactile sensing; characterization