Modular tendon-driven robot estimates own shape without external sensors
Tendon-Driven Continuum Robot with Modular Stiffness and In-Situ Self Pose Estimation
Robotics
Summary
Robots that bend smoothly like tentacles can be tricky to control and sense in tight spaces. This paper shows a new robot made of connected bendy joints that you can swap out to change its stiffness and shape. The robot can sense its own shape and position without using outside cameras or tracking devices, thanks to built-in magnetic sensors and smart software that learns how each joint moves. The researchers tested the robot in real environments and showed it adapts and knows its own pose by itself.
What this means in practice
- •For roboticists and automation engineers: Build adaptable continuum robots that reconfigure quickly and estimate their own shape without relying on external tracking systems.
- •For medical device developers: Design flexible surgical robots that measure their own pose inside confined environments without extra sensors, improving safety and deployment.
Authors
Guo Ning, Sue, Zheng Cao, Junzhe Hu, Xiangyun Bu, David Quinn, Tiancheng Wu, Zackory Erickson, Carmel Majidi
Abstract
Continuum robots enable smooth shape morphing and safe interaction in confined environments. However, most existing systems are task-specific and depend on external sensing infrastructure, limiting their adaptability and real-world deployment. This paper presents a self-contained modular continuum robotic platform that combines mechanical reconfigurability with onboard pose estimation. The robot is constructed from interchangeable continuum joints with analytically precomputed stiffness, allowing rapid assembly and direct programming of the robot shape. Proprioceptive sensing is achieved using magnetic sensors and a modular learning-based framework, where a single model is trained per joint and reused across configurations. The system is experimentally validated in real world, demonstrating self-sensing capabilities and adaptation without external tracking.