IEEE RA-L · 2023

Contour based object-compliant shape control

Ignacio Cuiral-Zueco, Yiannis Karayiannidis, Gonzalo López-Nicolás

I. Cuiral-Zueco and G. López-Nicolás are with Instituto de Investigación en Ingeniería de Aragón, Universidad de Zaragoza, Spain, ignaciocuiral@unizar.es, gonlopez@unizar.es.

Y. Karayiannidis is with Department of Automatic Control, Lund University, and is a member of the ELLIIT Strategic Research Area, yiannis@control.lth.se.

Cite this paper

I. Cuiral-Zueco, Y. Karayiannidis, and G. López-Nicolás, “Contour Based Object-Compliant Shape Control”, in IEEE Robotics and Automation Letters, vol. 8, no. 8, pp. 5164-5171, 2023, doi: 10.1109/LRA.2023.3292617.

Abstract

Shape control strategies seek to bring deformable objects towards a desired target shape. However, conventional methods focus on reaching the target shape without considering the extent to which the object is deformed during the control process. Control actions may generate unnecessary deformations and thus increase the possibility of object over-stressing and failure. In this paper we tackle the problem of vision-based object-compliant shape control (OCSC) with focus on 2D contours of texture-less objects. We propose a novel shape control framework that makes use of a multi-scale analysis of the contour's length and curvature to reduce the amount of deformation objects need to undergo during the shape control task. We evaluate our approach with simulations and experiments.

Video


Sections Overview


Acknowledgements

This work was supported via projects PID2021-124137OB-I00 and TED2021-130224B-I00 funded by MCIN/AEI/10.13039/501100011033, by ERDF A way of making Europe and by the European Union NextGenerationEU/PRTR. This work was also partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation.


Authors