# CaRoBio: 3D Cable Routing with a Bio-inspired Gripper Fingernail

arXiv: https://arxiv.org/abs/2508.09558 · HTML: https://mlsanity.com/paper/2508.09558/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 19 references that are in the map.
- First posted on arXiv 13 Aug 2025.

## Builds on (most cited first, 7 of 7)

- [Multi-Stage Cable Routing through Hierarchical Imitation Learning](https://mlsanity.com/paper/2307.08927/) (2023)
- [Offline-Online Learning of Deformation Model for Cable Manipulation with Graph Neural Networks](https://mlsanity.com/paper/2203.15004/) (2022)
- [Cable Routing and Assembly using Tactile-driven Motion Primitives](https://mlsanity.com/paper/2303.11765/) (2023)
- [Contact-aware Shaping and Maintenance of Deformable Linear Objects With Fixtures](https://mlsanity.com/paper/2307.10153/) (2023)
- [Feel the Tension: Manipulation of Deformable Linear Objects in Environments with Fixtures using Force Information](https://mlsanity.com/paper/2310.06424/) (2023)
- [In-Hand Following of Deformable Linear Objects Using Dexterous Fingers with Tactile Sensing](https://mlsanity.com/paper/2403.12676/) (2024)
- [Assembly of randomly placed parts realized by using only one robot arm with a general parallel-jaw gripper](https://mlsanity.com/paper/1909.08862/) (2019)

## Abstract (arXiv)

The manipulation of deformable linear flexures has a wide range of applications in industry, such as cable routing in automotive manufacturing and textile production. Cable routing, as a complex multi-stage robot manipulation scenario, is a challenging task for robot automation. Common parallel two-finger grippers have the risk of over-squeezing and over-tension when grasping and guiding cables. In this paper, a novel eagle-inspired fingernail is designed and mounted on the gripper fingers, which helps with cable grasping on planar surfaces and in-hand cable guiding operations. Then we present a single-grasp end-to-end 3D cable routing framework utilizing the proposed fingernails, instead of the common pick-and-place strategy. Continuous control is achieved to efficiently manipulate cables through vision-based state estimation of task configurations and offline trajectory planning based on motion primitives. We evaluate the effectiveness of the proposed framework with a variety of cables and channel slots, significantly outperforming the pick-and-place manipulation process under equivalent perceptual conditions. Our reconfigurable task setting and the proposed framework provide a reference for future cable routing manipulations in 3D space.