Separating Entangled Workpieces in Random Bin Picking using Deep Reinforcement Learning

Moosmann, Marius; Kulig, Marco; Spenrath, Felix; Mönnig, Manuel; Roggendorf, Simon; Petrovic, Oliver; Bormann, Richard; Huber, Marco F.

doi:10.1016/j.procir.2021.11.148

Cited by 14 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this problem, Leao et al [16] proposed a method to pick up soft tubes by fitting shape primitives to clutter. Moosmann et al [13] proposed to first estimate the 6D pose of the target and then leverage reinforcement learning to plan separation motion. However, these approaches require prior knowledge of the object shape or model.…”

Section: A Industrial Bin Pickingmentioning

confidence: 99%

“…However, these approaches use partial visual observation or simple *Correspond to: chou@hlab.sys.es.osaka-u.ac.jp geometrical features such as edges, making it challenging to be adopted in dense clutter. Other studies use pose estimation to evaluate the entanglement level for each object [13], [14]. Such a paradigm relies on the full knowledge of the objects and may suffer from cumulative perception errors due to heavy occlusion or self-occlusion of an individual complexshaped object.…”

Section: Introductionmentioning

confidence: 99%

“…The entanglement detection naturally leads to the necessity for exploring separation strategies. Studies have proposed tilting the gripper to discard the entangled objects [16] or dragging the entangled object out of the clutter [13]. However, these object-specific strategies require prior knowledge of objects and may be insufficient for objects with different geometries.…”

Section: Introductionmentioning

confidence: 99%

“…However, these object-specific strategies require prior knowledge of objects and may be insufficient for objects with different geometries. Additionally, the aforementioned learning-based approaches rely on simulated supervision [13] or verifying the entanglement by simulated execution [11]. They do not provide any general criteria for entanglement in cluttered environments.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning Efficient Policies for Picking Entangled Wire Harnesses: An Approach to Industrial Bin Picking

Zhang

Domae

Wan

et al. 2023

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Industrial bin picking for tangled-prone objects requires the robot to either pick up untangled objects or perform separation manipulation when the bin contains no isolated objects. The robot must be able to flexibly perform appropriate actions based on the current observation. It is challenging due to high occlusion in the clutter, elusive entanglement phenomena, and the need for skilled manipulation planning. In this paper, we propose an autonomous, effective and general approach for picking up tangled-prone objects for industrial bin picking. First, we learn PickNet -a network that maps the visual observation to pixel-wise possibilities of picking isolated objects or separating tangled objects and infers the corresponding grasp. Then, we propose two effective separation strategies: Dropping the entangled objects into a buffer bin to reduce the degree of entanglement; Pulling to separate the entangled objects in the buffer bin planned by PullNet -a network that predicts position and direction for pulling from visual input. To efficiently collect data for training PickNet and PullNet, we embrace the self-supervised learning paradigm using an algorithmic supervisor in a physics simulator. Realworld experiments show that our policy can dexterously pick up tangled-prone objects with success rates of 90%. We further demonstrate the generalization of our policy by picking a set of unseen objects. Supplementary material, code, and videos can be found at https://xinyiz0931.github.io/tangle.

show abstract

Section: A Industrial Bin Pickingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning Efficient Policies for Picking Entangled Wire Harnesses: An Approach to Industrial Bin Picking

Zhang

Domae

Wan

et al. 2023

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Existing solutions to such tasks [1][2][3][4][5][6] are commonly comprised of a task-agnostic grasp (TAG) pose estimator and a manipulation trajectory generator. They works rely on TAG poses and simple straight-up lifting motions that cannot guarantee separation in all situations.…”

Section: Introductionmentioning

confidence: 99%

Gam: General Affordance-Based Manipulation for Contact-Rich Object Disentangling Tasks

Yang,

Wu,

Lai

et al. 2023

Preprint

View full text Add to dashboard Cite

Using Meta-Learning to Reduce the Effort of Training New Workpiece Geometries for Entanglement Detection in Bin-Picking Applications

Moosmann,

Bleifuß,

Rosport

et al. 2023

Arena2036

View full text Add to dashboard Cite

In this paper, we introduce a scaling method for the training of neural networks used for entanglement detection in Bin-Picking. In the Bin-Picking process of complex-shaped and chaotically stored objects, entangled workpieces are a common source of problems. It has been shown that deep neural networks, which are trained using supervised learning, can be used to detect entangled workpieces. However, this strategy requires time-consuming data generation and an additional training process when adapting to previously unseen geometries. To solve this problem, we analyze and compare several Meta-Learning techniques like Reptile, MAML and TAMS for their feasibility as a scaling method for the entanglement detection. These methods search for a strongly generalized model for entanglement detection by learning from the training process of various workpieces with different geometries. Using this generalized model for entanglement detection as initialization helps to increase the learning success with only few training epochs and reduces the required amount of data and therefore the setup effort significantly.

show abstract

Separating Entangled Workpieces in Random Bin Picking using Deep Reinforcement Learning

Cited by 14 publications

References 14 publications

Learning Efficient Policies for Picking Entangled Wire Harnesses: An Approach to Industrial Bin Picking

Learning Efficient Policies for Picking Entangled Wire Harnesses: An Approach to Industrial Bin Picking

Gam: General Affordance-Based Manipulation for Contact-Rich Object Disentangling Tasks

Using Meta-Learning to Reduce the Effort of Training New Workpiece Geometries for Entanglement Detection in Bin-Picking Applications

Contact Info

Product

Resources

About