Interactive singulation of objects from a pile

Chang, Lillian Y.; Smith, Joshua R.; Fox, Dieter

doi:10.1109/icra.2012.6224575

Cited by 96 publications

(76 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, the robot is provided with a priori knowledge of the object type (a Lego block). Our work extends [16], [17] by considering more complex objects (unknown, natural, and complex shapes), to consider more complex clutter (piles), and by introducing learning to guide the interaction.…”

Section: B Learning Manipulation Expertisementioning

confidence: 99%

“…Recent work on pile manipulation [16], [17] relies on hardcoded heuristics for removing objects from a pile. In [16], flat objects are clustered together and the robot pokes objects until individual objects are singulated and can be picked up.…”

Section: B Learning Manipulation Expertisementioning

confidence: 99%

“…In [16], flat objects are clustered together and the robot pokes objects until individual objects are singulated and can be picked up. And in [17], Lego blocks are singulated and removed.…”

Section: B Learning Manipulation Expertisementioning

confidence: 99%

See 2 more Smart Citations

Perceiving, learning, and exploiting object affordances for autonomous pile manipulation

et al. 2014

View full text Add to dashboard Cite

Abstract-Autonomous manipulation in unstructured environments presents roboticists with three fundamental challenges: object segmentation, action selection, and motion generation. These challenges become more pronounced when unknown manmade or natural objects are cluttered together in a pile. We present an end-to-end approach to the problem of manipulating unknown objects in a pile, with the objective of removing all objects from the pile and placing them into a bin. Our robot perceives the environment with an RGB-D sensor, segments the pile into objects using non-parametric surface models, computes the affordances of each object, and selects the best affordance and its associated action to execute. Then, our robot instantiates the proper compliant motion primitive to safely execute the desired action. For efficient and reliable action selection, we developed a framework for supervised learning of manipulation expertise. We conducted dozens of trials and report on several hours of experiments involving more than 1500 interactions. The results show that our learning-based approach for pile manipulation outperforms a common sense heuristic as well as a random strategy, and is on par with human action selection.

show abstract

Section: B Learning Manipulation Expertisementioning

confidence: 99%

Section: B Learning Manipulation Expertisementioning

confidence: 99%

See 1 more Smart Citation

Perceiving, learning, and exploiting object affordances for autonomous pile manipulation

et al. 2014

View full text Add to dashboard Cite

show abstract

“…For example, in [10], active curiositydriven manipulation is used when learning the appearance of objects. Chang [4] presents a framework that can singulate objects contained in a pile through multiple interactions with the pile by pushing objects using a robot manipulator. In this framework, a decision module distinguishes between a single-item pile and a multiple-item pile, while a perception module combined with pushing actions accumulates evidence of items over multiple interactions to reduce the number of grasp errors.…”

Section: Related Workmentioning

confidence: 99%

Interactive perception based on Gaussian Process classification for house-hold objects recognition & sorting

Khan

Sun

Aragon-Camarasa

et al. 2016

2016 IEEE International Conference on Robotics and Biomimetics (ROBIO)

View full text Add to dashboard Cite

Abstract-We present an interactive perception model for object sorting based on Gaussian Process (GP) classification that is capable of recognizing objects categories from point cloud data. In our approach, FPFH features are extracted from point clouds to describe the local 3D shape of objects and a Bag-of-Words coding method is used to obtain an objectlevel vocabulary representation. Multi-class Gaussian Process classification is employed to provide and probable estimation of the identity of the object and serves a key role in the interactive perception cycle -modelling perception confidence. We show results from simulated input data on both SVM and GP based multi-class classifiers to validate the recognition accuracy of our proposed perception model. Our results demonstrate that by using a GP-based classifier, we obtain true positive classification rates of up to 80%. Our semi-autonomous object sorting experiments show that the proposed GP based interactive sorting approach outperforms random sorting by up to 30% when applied to scenes comprising configurations of household objects.

show abstract

“…In a subsequent stage, movement and object membership can be estimated for each of these parts. Alternative approaches use algorithms such as iterative closest point (ICP) to determine whether the tracked point cloud is a single rigid object [12,31], or estimate the movement of trackable visual features [13,16,25,[31][32][33].…”

Section: B Interactive Perception For Object Segmentationmentioning

confidence: 99%

Probabilistic Segmentation and Targeted Exploration of Objects in Cluttered Environments

2014

View full text Add to dashboard Cite

Abstract-Creating robots that can act autonomously in dynamic, unstructured environments requires dealing with novel objects. Thus, an off-line learning phase is not sufficient for recognizing and manipulating such objects. Rather, an autonomous robot needs to acquire knowledge through its own interaction with its environment, without using heuristics encoding human insights about the domain. Interaction also allows information that is not present in static images of a scene to be elicited. Out of a potentially large set of possible interactions, a robot must select actions that are expected to have the most informative outcomes to learn efficiently. In the proposed bottom-up, probabilistic approach, the robot achieves this goal by quantifying the expected informativeness of its own actions in information-theoretic terms. We use this approach to segment a scene into its constituent objects. We retain a probability distribution over segmentations. We show that this approach is robust in the presence of noise and uncertainty in real-world experiments. Evaluations show that the proposed information-theoretic approach allows a robot to efficiently determine the composite structure of its environment. We also show that our probabilistic model allows straightforward integration of multiple modalities, such as movement data and static scene features. Learned static scene features allow for experience from similar environments to speed up learning for new scenes.

show abstract

Interactive singulation of objects from a pile

Cited by 96 publications

References 16 publications

Perceiving, learning, and exploiting object affordances for autonomous pile manipulation

Perceiving, learning, and exploiting object affordances for autonomous pile manipulation

Interactive perception based on Gaussian Process classification for house-hold objects recognition & sorting

Probabilistic Segmentation and Targeted Exploration of Objects in Cluttered Environments

Contact Info

Product

Resources

About