Fiber sensor tracks 3D shapes with tiny, laser-cut parts

A reel-to-reel femtosecond laser writing process now enables the creation of three-dimensional shape sensors within standard single-mode optical fibers. Researchers densely inscribed “misplaced orthogonal eccentric scatterers,” or MOESs, into the cladding of these fibers, a location distinct from the core, to build a miniaturized system for high-precision shape detection. Each MOES element acts as a curvature-dependent light scatterer, spatially encoding 3D deformation information. The intensity-domain reconstruction algorithm also exhibits strong immunity to environmental perturbations, opening possibilities for fiber sensors in unstructured and dynamic environments.

Femtosecond Laser Writing Fabricates MOES Arrays in Single-Mode Fiber

A manufacturing process utilizing a continuous, moving fiber allows for the dense creation of three-dimensional shape sensors, avoiding the painstaking assembly typically associated with high-precision instrumentation. Researchers detailed a “reel-to-reel femtosecond laser writing process” capable of fabricating arrays of uniquely designed structures within optical fibers, offering a pathway to scalable and potentially lower-cost sensor production. This method moves beyond traditional fabrication techniques that often rely on manual alignment and placement of individual components.

This precise positioning within the cladding, rather than the fiber’s core, is a key distinction from previous approaches to fiber optic sensing. The deliberate arrangement of these elements spatially encodes information about the fiber’s three-dimensional shape, all within the confines of a single fiber core. Experimental validation confirmed the ability of these orthogonally distributed eccentric scatterers to measure two-dimensional bending at a single point, a crucial step towards full 3D deformation reconstruction.

The team successfully demonstrated reconstruction of three-dimensional deformation using a fiber containing 36 of these precisely positioned scatterers. Importantly, the reconstruction algorithm operates in the intensity-domain, a design choice that provides strong immunity to environmental perturbations because the system relies on changes in light intensity, rather than more subtle spectral shifts, making it less susceptible to environmental factors like temperature fluctuations or strain.

The ability to fabricate these MOES arrays within a standard single-mode fiber simplifies the sensing system, eliminating the need for complex multi-channel demodulation equipment and minimizing the overall device size. This approach, the researchers suggest, opens opportunities for highly integrated and simplified sensing solutions applicable to a wide range of complex deformation monitoring requirements, particularly in environments where size and robustness are paramount.

Single-Fiber Sensing Overcomes Size and Resolution Trade-offs

Researchers are developing a new approach to three-dimensional shape sensing, moving beyond bulky multi-fiber systems to a single-fiber design fabricated with a scalable process. This innovation addresses a critical limitation in existing fiber optic sensors; traditionally, increasing device size was necessary to achieve higher spatial resolution, hindering integration into increasingly compact applications like soft robotics and minimally invasive surgery.

Rayleigh Scattering Measures Curvature in MOES-Based System

The fabrication of highly sensitive three-dimensional shape sensors now benefits from a manufacturing process typically reserved for high-volume production, rather than painstaking laboratory assembly. A team has demonstrated a method for creating these sensors using a scalable process involving densely inscribed features within standard optical fibers, potentially lowering costs and accelerating deployment in applications like soft robotics and medical devices. This advance centers on the creation of an array of “misplaced orthogonal eccentric scatterers,” or MOESs, positioned within the fiber’s cladding, the layer surrounding the light-carrying core, rather than disrupting the core itself.

These MOESs function as curvature-dependent Rayleigh scatterers, meaning they alter how light is dispersed when the fiber bends. The team utilized a technique to fabricate these arrays, enabling dense inscription of the MOESs along the fiber length. This contrasts with previous methods that required meticulous, individual placement of sensing elements, a process that limited scalability and increased expense. This approach, they believe, “defines a new paradigm for compact, cost-effective, and high-fidelity fiber sensors,” opening possibilities for broader adoption in fields demanding precise and reliable shape sensing.

Intensity-Domain Reconstruction Enables Robust Shape Sensing

The challenge of creating truly compact and reliable three-dimensional shape sensors has long hinged on balancing spatial resolution with device size; current systems often rely on bulky multi-fiber or multi-core designs. The core innovation lies in how this sensor interprets deformation. Rather than relying on wavelength shifts or complex optical interference patterns, the system operates in the intensity-domain to determine the fiber’s shape. This method, the researchers report, offers strong immunity to environmental perturbations.

Experimental results demonstrate the system’s ability to achieve high-fidelity reconstruction of both two-dimensional bending and full three-dimensional deformation. This simplification, combined with the scalable fabrication process, positions the technology for broader adoption in fields like soft robotics and minimally invasive medical devices, where compact and robust sensing is paramount.

Limitations of Existing OFDR and FBG Technologies

Current optical fiber sensors, while increasingly used for high-precision shape detection in areas like soft robotics and medical devices, historically faced a trade-off between sensor size and the detail of spatial resolution they could achieve. Traditional approaches relying on distributed optical frequency domain reflectometry (OFDR) and fiber Bragg gratings (FBGs) often necessitated multi-channel fiber bundles or multi-core fibers to achieve continuous deformation measurement, increasing complexity and bulk.

Commercial products utilizing these basic principles have begun to appear for medical and robotic applications, yet limitations remain as applications demand greater precision and integration. The core challenge lies in shrinking sensor dimensions while maintaining, or even improving, the ability to accurately map complex three-dimensional shapes. These established OFDR and FBG techniques, despite their strengths, struggle to deliver the compactness needed for advanced soft robotic systems.

While OFDR offers near-unlimited sensing points along a fiber’s length, it lacks wavelength markers, preventing it from distinguishing two-dimensional bending angles. Similarly, FBGs, which use wavelength-based identification, are constrained by the spectral bandwidth of the light source; current studies report fewer than ten gratings per channel with unsatisfactory spectral quality. This limitation restricts the density of measurable points and the overall fidelity of deformation reconstruction.

Researchers addressed this by exploring optical switching to sequentially interrogate multiple channels, but this compromises speed and real-time feedback capabilities, a critical factor for dynamic applications. An alternative approach, Rayleigh-signature domain multiplexing, uses cross-correlation to separate multi-channel scattering signals, but its reliability in noisy or rapidly changing environments remains a concern. These shortcomings prompted the development of new sensing schemes capable of simultaneously achieving high spatial resolution, robust feature recognition, and compact architectures.

Recent progress in femtosecond laser direct writing has enabled the creation of specialized devices like cladding FBGs and Mach-Zehnder interferometers, allowing for one- or two-dimensional bending measurement within a single fiber. However, achieving continuous two-dimensional bending measurements, essential for reconstructing curves, remained a significant hurdle. The ideal solution, as identified by the developers of this new system, is a single-channel fiber-optic deformation sensor that combines the wavelength-specific feature recognition of FBGs with the dense spatial sampling of OFDR.

The new technology aims to overcome these limitations by employing a spatial encoding strategy, mapping three-dimensional deformation to the deliberate arrangement of eccentric scatterers within the fiber cladding. This approach, they report, offers a pathway to richer deformation information extraction with minimal data input, reducing costs and accelerating signal processing.

Soft Robotics Drives Demand for High-Precision Fiber Sensors

Demand for increasingly sophisticated sensing systems is being fueled by rapid advancements in soft robotics, and a new approach to fiber optic shape sensing offers a potential solution to longstanding limitations in spatial resolution and device size. Experimental validation has demonstrated high-fidelity reconstruction of both 2D bending and full 3D deformation, indicating a significant step toward more integrated and simplified sensing solutions. The team’s design prioritizes compactness, aiming to optimize space utilization within increasingly complex soft robotic systems.

The system utilizes optical frequency domain reflectometry (OFDR) as a single-channel spatial readout tool, while the deformation encoding itself is achieved through the spatial configuration of the scatterers, rather than relying on wavelength-domain discrimination. The team successfully demonstrated reconstruction of three-dimensional deformation using a fiber containing 36 of these precisely positioned scatterers, and the intensity-domain reconstruction algorithm exhibits strong immunity to environmental perturbations.

MOES Approach Combines FBG and OFDR Advantages

Researchers at the forefront of soft robotics are refining methods for precise spatial awareness within increasingly compact systems, and a new approach leverages a scalable fabrication technique to create highly sensitive three-dimensional shape sensors. This fabrication method moves beyond painstaking assembly, offering a path toward cost-effective manufacturing of complex sensors.

Unlike existing fiber optic technologies that often require multi-channel fiber bundles or multi-core fibers, this system operates with a single fiber, optimizing space utilization and simplifying integration into delicate robotic structures. This differs from traditional fiber optic sensors that rely on wavelength-based identification, which can be limited by spectral bandwidth.