2016
DOI: 10.48550/arxiv.1604.03639
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

A Summary of Team MIT's Approach to the Amazon Picking Challenge 2015

Abstract: The Amazon Picking Challenge (APC) [1], held alongside the International Conference on Robotics and Automation in May 2015 in Seattle, challenged roboticists from academia and industry to demonstrate fully automated solutions to the problem of picking objects from shelves in a warehouse fulfillment scenario. Packing density, object variability, speed, and reliability are the main complexities of the task. The picking challenge serves both as a motivation and an instrument to focus research efforts on a specifi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0

Year Published

2017
2017
2022
2022

Publication Types

Select...
10

Relationship

1
9

Authors

Journals

citations
Cited by 23 publications
(16 citation statements)
references
References 9 publications
0
16
0
Order By: Relevance
“…When planning suction grasps, existing approaches either directly propose grasps on a point cloud of the scene using heuristic methods or sample a range of candidate grasps and rank them using a quality metric. For the former, common approaches are grasping near estimated centroids of flat surfaces [25], grasping along inward surface normals towards an object centroid [10], or pushing objects from the top or side until a suction seal is formed [8]. In the latter case, Domae et al [7] use a geometric model that assesses planarity by convolving a contact template with the image.…”
Section: B Suction Graspingmentioning
confidence: 99%
“…When planning suction grasps, existing approaches either directly propose grasps on a point cloud of the scene using heuristic methods or sample a range of candidate grasps and rank them using a quality metric. For the former, common approaches are grasping near estimated centroids of flat surfaces [25], grasping along inward surface normals towards an object centroid [10], or pushing objects from the top or side until a suction seal is formed [8]. In the latter case, Domae et al [7] use a geometric model that assesses planarity by convolving a contact template with the image.…”
Section: B Suction Graspingmentioning
confidence: 99%
“…MaskRCNN is a neural network used for object segmentation [21]. The dual-arm system uses a customised version of MaskRCNN written in Tensorflow 5 . The version of Tensorflow used requires older GPU drivers not supported officially by Ubuntu 18.…”
Section: Maskrcnnmentioning
confidence: 99%
“…Since the Amazon Picking Challenge 2015, a variety of approaches to a more general bin-picking problem had been proposed [17]- [19]. In [20], authors discussed their observations and lessons drawn from a survey conducted among participating teams in the challenge.…”
Section: Related Workmentioning
confidence: 99%