Ionic hydrogel sensors for low-latency high-accuracy 3D reconstruction of soft robotic endoscope

Continuous shape sensing remains a major barrier to deploying soft robotic endoscopes, because existing solutions often compromise structural compliance, require expensive instrumentation, or provide only limited embedded feedback. Here, we report an integrated soft sensing system based on an array of ionic hydrogel strain sensors, enabling low-latency, high-accuracy three-dimensional shape reconstruction of a soft endoscope tip. Optimization across NaCl concentration, applied voltage, and strain rate identifies ideal composition and operating conditions jointly balancing gauge factor, strain-rate independence, and mechanical robustness. Three optimized sensors are then integrated around the circumference of a soft pneumatic actuator, addressing adhesion, dehydration, and signal stability under actuation. Continuous three-dimensional shape reconstruction is achieved using a Frenet-Serret-based model, and a simple delay-compensation strategy reduces the relative tip-position error from 10.08% to 5.92% of the actuator arc length. This offers a low-cost, compliant, disposable-compatible route to proprioceptive sensing for minimally invasive intervention.