Inductance-Based Force Self-Sensing in Fiber-Reinforced Pneumatic Twisted-and-Coiled Actuators
Yunsong Zhang, Tianlin Li, Mingyang Yang, Feitian Zhang * · 2026
Read Inductance-Based Force Self-Sensing in Fiber-Reinforced Pneumatic Twisted-and-Coiled Actuators by Yunsong Zhang, Tianlin Li, Mingyang Yang, Feitian Zhang * (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18555.txt.
Introduction
Zhang, Li, Yang, and Zhang give a twisted-and-coiled soft actuator a sense of its own force. These pneumatic actuators are strong and compliant but show strong hysteresis. A nickel wire inside lets inductance track force with low hysteresis at fixed pressure. Length tracking, by contrast, stays highly hysteretic. A model plus a hybrid observer using a Kalman filter estimates force and state. Lab tests match load cell readings well, even as loads change during motion.
Catalog record
| CREATOR | Yunsong Zhang, Tianlin Li, Mingyang Yang, Feitian Zhang * |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37264 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.18555
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Inductance-Based Force Self-Sensing in Fiber-Reinforced Pneumatic Twisted-and-Coiled Actuators Yunsong Zhang, Tianlin Li, Mingyang Yang, Feitian Zhang * Abstract Fiber-reinforced pneumatic twisted-and-coiled actuators (FR-PTCAs) offer high power density and compliance but their strong hysteresis and lack of intrinsic proprioception limit effective closed-loop control. This paper presents a self-sensing FR-PTCA integrated with a conductive nickel wire that enables intrinsic force estimation and indirect displacement inference via inductance feedback. Experimental characterization reveals that the inductance of the actuator exhibits a deterministic, low-hysteresis inductance-force relationship at constant pressures, in contrast to the strongly hysteretic inductance-length behavior. Leveraging this property, this paper develops a parametric self-sensing model and a nonlinear hybrid observer that integrates an Extended Kalman Filter (EKF) with constrained optimization to resolve the ambiguity in the inductance-force mapping and estimate actuator states. Experimental results demonstrate that the proposed approach achieves force estimation accuracy comparable to that of external load cells and maintains robust performance under varying load conditions. I Introduction Artificial muscles have attracted significant attention in the field of robotics due to their inherent compliance and high power-to-weight ratios [ 23 ] . Among various actuation technologies, Twisted and Coiled Actuators (TCAs) have emerged as a prominent class owing to their high work density and low manufacturing cost [ 5 , 23 , 21 ] . However, conventional TCAs are typically driven by electrothermal actuation, which leads to limited energy efficiency and slow dynamic responses [ 5 , 1 ] . To address these limitations, fluid-driven variants, such as the Cavatappi artificial muscle, have been proposed [ 7 , 6 ] . More recently, Weissman et al. introduced the Fiber-Reinforced Pneumatic TCA (FR-PTCA), which employs fiber reinforcement to enhance structural anisotropy [ 19 ] . This design significantly improves energy efficiency (exceeding 19\% ) and achieves up to 70\% contraction strain at relatively low pressures, making the FR-PTCA a promising actuation solution for high-performance soft robotic systems. Despite these mechanical advances, the practical deployment of FR-PTCAs remains limited by the lack of intuitive modeling and sensing frameworks for real-time control [ 6 , 19 ] . Unlike thermal TCAs, for which linear time-invariant models have been derived to facilitate feedback control [ 20 , 14 ] , pneumatic TCAs currently lack a unified modeling approach. This challenge becomes particularly pronounced in soft continuum robots, where accurate state estimation is difficult. Conventional rigid sensors, such as encoders or external motion capture systems, introduce additional mass and mechanical stiffness, which compromises the compliance adaptability of soft robotic systems [ 10 ] . Consequently, the development of FR-PTCAs with integrated self-sensing capabilities, whereby the actuator simultaneously functions as a sensor, is critical. Such capability would enable accurate state estimation and closed-loop control while preserving the fundamental compliance of robotic systems [ 8 , 14 , 11 ] . Fig. 1: Conceptual overview of the proposed inductive self-sensing framework for FR-PTCAs. A helically wrapped conductive nickel wire enables intrinsic inductive sensing without external rigid sensors. While the inductance-length relationship exhibits strong hysteresis, the inductance-force mapping remains deterministic and low-hysteresis at constant pressures. Leveraging this property, a nonlinear hybrid observer processes inductance measurements to enable real-time estimation of actuator force and internal states. Self-sensing refers to the capability of extracting actuator state information by exploiting its intrinsic electrical or mechanical properties [ 4 ] . This concept has been widely investigated in artificial muscle systems. For example, piezoelectric actuators utilize capacitance for position estimation [ 4 ] . Thermal TCAs typically rely on electrical resistance or impedance to infer temperature and strain [ 15 , 2 , 8 , 17 ] . However, these approaches often suffer from strong thermal coupling, temperature drift, and hysteresis, which complicate accurate state estimation [ 17 , 18 , 13 ] . Regarding pneumatic artificial muscles (PAMs), several approaches have been explored, including inductive sensing using embedded springs [ 3 ] , capacitance-based sensing via dielectric elastomer sensors [ 9 ] , and magnetic field-based sensing for displacement and force estimation [ 12 ] . Sharing the twisted helical geometry of thermal TCAs and the fluid-driven actuation of PAMs, FR-PTCAs represent a unique technological intersection. By inherently bypassing the thermal complications present in traditional TCAs, FR-PTCAs offer a distinct opportunity to exploit inductive sensing for stable, high-fidelity self-sensing. Building upon the structural assumptions proposed by Weissman et al. [ 19 ] , namely the inextensibility of the reinforcement fiber and the no-slip condition between the fiber and the tube, the FR-PTCA can be viewed as a nested double-helix structure comprising a macroscopic helical coil and a microscopic twisted fiber trajectory. Under these kinematic constraints, we hypothesize that the integration of conductive reinforcing fibers maintains tight coupling with actuator deformation. As illustrated in Fig. 1 , this configuration enables the actuator inductance to couple preferentially with internal tension and output force rather than with the macroscopic actuator length. Experimental observations suggest that actuator inductance exhibits a deterministic and low-hysteresis relationship with the output force at constant pressures, in contrast to the strongly hysteretic inductance-length relationship. This physical property provides the basis for a reliable force self-sensing mechanism and motivates the observer design developed in this work. This paper presents a self-sensing FR-PTCA and a unified framework for its fabrication, modeling and control. We replace the conventional nylon monofilament with a conductive nickel wire, endowing the actuator with intrinsic inductive sensing capability while preserving its mechanical compliance. Experimental characterization reveals a deterministic, low-hysteresis relationship between actuator inductance and output force at constant pressures. A dynamic model is established to characterize the interplay between force, pressure, and length. Leveraging this property, we develop a parametric self-sensing model and a nonlinear hybrid observer that integrates an EKF with constrained optimization to estimate actuator states. Finally, inductance-based feedback control is implemented and experimentally evaluated against open-loop and sensor-based control strategies. The results demonstrate that the proposed self-sensing method achieves force estimation accuracy comparable to external load cells and significantly improves tracking accuracy under varying load conditions. The main contributions of this paper are threefold. First, the design and fabrication of a self-sensing FR-PTCA are presented, providing intrinsic proprioception while preserving mechanical compliance. Second, dynamic and self-sensing models are established to characterize the low-hysteresis coupling between the inductance, internal pressure, and output force. Third, a hybrid EKF-optimization observer is developed to facilitate real-time state estimation by resolving non-monotonicity in the inductance mapping. This integrated framework provides a robust solution for the control of soft robotic systems without the need for external instrumentation. II Design and Fabrication To enable inductance-based self-sensing, the fabrication protocol of the standard FR-PTCA [ 19 ] was modified by replacing the conventional reinforcing fiber with a conductive nickel wire. The fabrication process consists of six stages: 1) precursor preparation, 2) drawing, 3) conductive fiber reinforcement, 4) twisting, 5) coiling, and 6) annealing. As illustrated in the schematic of the custom fabrication platform in Fig. 2 , the apparatus incorporates two rotary stepper motors (Motors A and B) to apply torque, along with two linear stepper motors equipped with guide rails (Motors C and D) to control axial displacement and tension. Fig. 2: Fabrication process of the self-sensing FR-PTCA. Schematics of the fabrication platform and its stages: (a) precursor preparation and drawing, (b) conductive