2016
DOI: 10.48550/arxiv.1603.02199
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collection

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
81
0
1

Year Published

2016
2016
2024
2024

Publication Types

Select...
3
2
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 44 publications
(82 citation statements)
references
References 0 publications
0
81
0
1
Order By: Relevance
“…In bin-picking, objects are located in a pile and are often small and light. Methods that target bin-picking [1]- [3], [8], [9] benefit from the pile structure as objects have more stable equilibriuma and as such collisions with obstacles does not jeopardise the pile structure. In comparison, methods that target grasping in structured clutter [4]-where objects are mostly larger, heavier, and packed together-need to avoid all collisions as unintended contact between the robot and objects can easily tip objects over and thus change the scene structure.…”
Section: B Grasping In Cluttermentioning
confidence: 99%
See 3 more Smart Citations
“…In bin-picking, objects are located in a pile and are often small and light. Methods that target bin-picking [1]- [3], [8], [9] benefit from the pile structure as objects have more stable equilibriuma and as such collisions with obstacles does not jeopardise the pile structure. In comparison, methods that target grasping in structured clutter [4]-where objects are mostly larger, heavier, and packed together-need to avoid all collisions as unintended contact between the robot and objects can easily tip objects over and thus change the scene structure.…”
Section: B Grasping In Cluttermentioning
confidence: 99%
“…In cluttered scenes, the most successful methods have resorted to the simper 4-DOF planar top-down grasps [1]- [3], [8] instead of full 6-DOF spatial grasps [4], [7], [22]. While 4-DOF grasps are simpler, the restriction on the arm motion they imply hinders the robot's ability to grasp specific objects [4], [7], especially with structured clutter.…”
Section: B Grasping In Cluttermentioning
confidence: 99%
See 2 more Smart Citations
“…End-to-end learning techniques, such as deep reinforcement learning algorithms, combine both processes to directly build policies [3], and can theoretically be used to learn new tasks and adapt to environmental variations in such open-ended scenarios. However, they require very large learning datasets, which is not suited to real robotics tasks where evaluating the outcome of actions or policies is slow and costly [4], except in the rare cases when many real robots can be used in parallel [3]. They are also notably slow to converge when reward states are sparse, which is often the case in tasks necessitating to reach some very specific goal states using only partial, high-dimensional and noisy obervations of the world.…”
Section: Introductionmentioning
confidence: 99%