aiqbench analysis
This research addresses a critical scalability challenge in soft robotics by enabling visual shape control across varying robot configurations without retraining models for each new module count. By decomposing whole-body images into local patches and leveraging geometric data augmentation, the method allows a single controller trained on one module to effectively manage serial robots with one to five modules. This approach significantly reduces data collection overhead and enhances adaptability to environmental changes and payload variations, marking a practical step toward more flexible soft robotic systems.
Module Number Adaptive Visual Shape Control for Serial Modular Soft Robots
arXiv:2608.29547v1 Announce Type: new Abstract: Image based shape control provides a simple means of controlling the whole body configuration of soft robots. However, existing data driven approaches are typically developed for fixed robot structures and require new control data when the number of modules changes. This paper presents a module number adaptive visual shape control method for serial modular soft pneumatic robots. A controller trained only on single module actuation shape data is reused for robots with one to five modules by decomposing whole body camera images into local module patches. A single common module segmenter localizes individual modules across all tested configurations, while the same local controller is applied to every extracted patch. Geometric data augmentation improves transferability to downstream modules, and a lightweight mask reconstruction network reconstructs a synthetically removed actuator mask channel. Experiments on physical robots demonstrate shape control across varying numbers of modules and under environmental changes and payload loading. The results show that single module control learning enables scalable whole body control without configuration specific control data collection.
Abstract:Image based shape control provides a simple means of controlling the whole body configuration of soft robots. However, existing data driven approaches are typically developed for fixed robot structures and require new control data when the number of modules changes. This paper presents a module number adaptive visual shape control method for serial modular soft pneumatic robots. A controller trained only on single module actuation shape data is reused for robots with one to five modules by decomposing whole body camera images into local module patches. A single common module segmenter localizes individual modules across all tested configurations, while the same local controller is applied to every extracted patch. Geometric data augmentation improves transferability to downstream modules, and a lightweight mask reconstruction network reconstructs a synthetically removed actuator mask channel. Experiments on physical robots demonstrate shape control across varying numbers of modules and under environmental changes and payload loading. The results show that single module control learning enables scalable whole body control without configuration specific control data collection.
9 pages, 7 figures. This articles has been submitted to a IEEE Journal
Robotics (cs.RO)
arXiv:2608.29547 [cs.RO]
(or arXiv:2608.29547v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2608.29547 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 274 words
