3D Reconstruction of Unknown Objects

Performing tasks in unstructured environments is a challenging robotic research field. In applications like object fine manipulation, the object's surface must be known to properly plan the coordinated movements of the robotic fingers. In the proposed algorithms, a robot equipped with a calibrated camera, mounted in an eye-in-hand configuration, follows some trajectories around the object to be reconstructed. Several images are acquired along these paths. After some standard image processing operations, an ellipsoid is virtually placed around the object, and it is sampled by points, in turn, interconnected through virtual springs and dampers. These points dynamically shrink toward the ellipsoid's centre and stop when intercepting the so-called visual hull. In this way, it is possible to quickly and accurately reconstruct the objects' surfaces in unstructured environments.

Proposed solutions:

• V. Lippiello, F. RUGGIERO, B. Siciliano, Floating visual grasp of unknown objects using an elastic reconstruction surface, in Robotics Research: The Fourteenth Inter- national Symposium, in Springer Tracts in Advanced Robotics 70, C. Pradalier, R. Siegwart and G. Hirzinger (Eds.), pp. 329-344, Springer, Heidelberg, D, 2011, DOI: 10.1007/978-3-642-19457-3_20.
• J.-T. Kim, F. RUGGIERO, V. Lippiello, B. Siciliano, Smoothed particle hydrodynamics-based viscous deformable object modelling, in Robot dynamic manipulation. Percep- tion of deformable objects and nonprehensile manipulation control, B. Siciliano and F. Ruggiero (eds.), vol. 144, pp. 73-102, Springer, Cham, 2022. DOI: 10.1007/978-3- 030-93290-9_3.
• V. Lippiello, F. RUGGIERO, Surface model Reconstruction of 3D objects from multiple views, 2009 IEEE International Conference on Robotics and Automation, Kobe, J, pp. 2400—2405, 2009, DOI: 10.1109/ROBOT.2009.5152652.