Adaptive Bendpoint Formation in Octopus-Inspired Soft Continuum Arms
cs.RO
Submitted: 2026-10-18
Updated: 2026-10-18
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
The gist: Octopuses exhibit a powerful form of embodied intelligence from a highly adaptable soft body.
Abstract
Octopuses exhibit a powerful form of embodied intelligence from a highly adaptable soft body. One key mechanism is the ability to instantiate bending points at any location along the arm, which enables performant and dexterous behaviours. Taking inspiration from the octopus for its dexterity, this study focuses on changing the curvature of a tendon-actuated arm by adaptively changing the anchoring end-point along the arm. An array of inflatable shells is integrated into the arm, where inflating a shell causes an increase in friction at the location, jamming the tendon and creating a temporary new anchoring point. The novel soft actuator does not break the continuity of the arm between the actuated segment and the loose segment. Experiments show that the actuated segment has a non-constant bell-shaped curvature which is comparable to that of a real octopus arm during a reaching motion. Compared to conventional single-tendon continuum manipulators, the proposed design offers an expanded workspace and enhanced dexterity through the ability to independently shape different regions of the arm through selective curvature tuning. At the same time, it avoids the mechanical complexity typically associated with multi-segment, multi-actuator systems.
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies