Automatic observation for 3D reconstruction of unknown objects using visual servoing

Walck, Guillaume; Drouin, M.

doi:10.1109/iros.2010.5649507

Cited by 9 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using scene exploration with embedded sensors to reconstruct and manipulate a 3D model of unknown objects is an approach taken by several researchers [6,52,103,75,59]. Walck et al [103] propose a method which automatically finds the position of the targeted object using a single eye-in-hand camera, captures multiple views of its shape using visual servoing, and models unknown objects using carved visual Surveying this literature, there remains a need in the robotics community to develop techniques to reconstruct 3D models of articulated objects, particularly models that incorporate the non-visible portions of the objects for occlusion-aware sensing and manipulation.…”

Section: Object Manipulationmentioning

confidence: 99%

Occlusion-aware multi-view reconstruction of articulated objects for manipulation

Huang

Walker

Birchfield

2014

Robotics and Autonomous Systems

View full text Add to dashboard Cite

The goal of this research is to develop algorithms using multiple views to automatically recover complete 3D models of articulated objects in unstructured environments and thereby enable a robotic system to facilitate further manipulation of those objects. First, an algorithm called Procrustes-Lo-RANSAC (PLR) is presented.Structure-from-motion techniques are used to capture 3D point cloud models of an articulated object in two different configurations. Procrustes analysis, combined with a locally optimized RANSAC sampling strategy, facilitates a straightforward geometric approach to recovering the joint axes, as well as classifying them automatically as either revolute or prismatic. The algorithm does not require prior knowledge of the object, nor does it make any assumptions about the planarity of the object or scene.Second, with such a resulting articulated model, a robotic system is then able to manipulate the object either along its joint axes at a specified grasp point in order to exercise its degrees of freedom or move its end effector to a particular position even if the point is not visible in the current view. This is one of the main advantages of the occlusion-aware approach, because the models capture all sides of the object meaning that the robot has knowledge of parts of the object that are not visible in the current view. Experiments with a PUMA 500 robotic arm demonstrate the effectiveness of the approach on a variety of real-world objects containing both revolute and prismatic joints.ii Third, we improve the proposed approach by using a RGBD sensor (Microsoft Kinect) that yield a depth value for each pixel immediately by the sensor itself rather than requiring correspondence to establish depth. KinectFusion algorithm is applied to produce a single high-quality, geometrically accurate 3D model from which rigid links of the object are segmented and aligned, allowing the joint axes to be estimated using the geometric approach. The improved algorithm does not require artificial markers attached to objects, yields much denser 3D models and reduces the computation time.iii Dedication I dedicate this work to my beloved family who made all of this possible for their endless encouragement and faith.iv

show abstract

Section: Object Manipulationmentioning

confidence: 99%

Occlusion-aware multi-view reconstruction of articulated objects for manipulation

Huang

Walker

Birchfield

2014

Robotics and Autonomous Systems

View full text Add to dashboard Cite

show abstract

“…Following their work, various NBV methods have been proposed for applications in autonomous robotic operation. [13][14][15][16][17][18][19] Walck and Drouin 13 assumed an a priori knowledge of target object's location and surrounding environment and proposed an NBV structure for actively observing a single object using a monocular camera. A non-model NBV method using image boundaries as primary cues was introduced by Kriegel et al 14 A method regarding humanoid active perception of a grasped object has been developed, 15 in which the viewing pose is planned in a greedy manner that is similar to the NBV structure.…”

Section: Introductionmentioning

confidence: 99%

Active manipulator motion planning for planetary landform awareness

Leng

Meng

et al. 2019

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

This paper presents an active motion planning approach for a robotic manipulator operating in planetary surface exploration missions. A monocular camera is employed to examine the visually salient regions in an image, from which landforms of potential interests are extracted, and two key metrics are established to evaluate the information richness of the landforms. A next-best view manipulator motion planning is proposed, in which motions of the manipulator are actively planned to reach a better viewpoint to view the target landform, the safety of such operation is guaranteed by real-time estimating the relative distance using ORriented Brief (ORB)–simultaneous localization and mapping. The proposed active motion planning method is validated by an experimental trial, results from which demonstrate that it is able to safely acquire a better viewing of a just-detected planetary landform with full autonomy.

show abstract

Contour-based next-best view planning from point cloud segmentation of unknown objects

Monica

Aleotti

2017

Auton Robot

View full text Add to dashboard Cite

Automatic observation for 3D reconstruction of unknown objects using visual servoing

Cited by 9 publications

References 10 publications

Occlusion-aware multi-view reconstruction of articulated objects for manipulation

Occlusion-aware multi-view reconstruction of articulated objects for manipulation

Active manipulator motion planning for planetary landform awareness

Contour-based next-best view planning from point cloud segmentation of unknown objects

Contact Info

Product

Resources

About