Leveraging Contact Forces for Learning to Grasp

Merzic, Hamza; Bogdanovic, Miroslav; Kappler, Daniel; Righetti, Ludovic; Bohg, Jeannette

doi:10.48550/arxiv.1809.07004

Search citation statements

Order By: Relevance

Paper Sections

Select...

Introduction2

A Stabilization Of the Linear Complementarity System1

A Cart-pole With Soft Walls1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2020

Publication Types

Select...

Other2

Relationship

Self Cite0

Independent2

Authors

Journals

Cited by 2 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Definition 5, it follows that λ(t) ∈ SOL(Ex + c, F ) and (x(t), λ(t)) ∈ Γ SOL (E, F, c) for almost all t ≥ 0. The result follows from (14).…”

Section: A Stabilization Of the Linear Complementarity Systemmentioning

confidence: 75%

See 1 more Smart Citation

Stabilization of Complementarity Systems via Contact-Aware Controllers

Aydinoglu¹,

Sieg²,

Preciado³

et al. 2020

Preprint

View full text Add to dashboard Cite

“…From Definition 5, it follows that λ(t) ∈ SOL(Ex + c, F ) and (x(t), λ(t)) ∈ Γ SOL (E, F, c) for almost all t ≥ 0. The result follows from (14).…”

Section: A Stabilization Of the Linear Complementarity Systemmentioning

confidence: 75%

“…[12], [13]). Other recent methods incorporate tactile sensors within deep learning frameworks, though offer no guarantees on performance or stability [14], [15].…”

Section: Introductionmentioning

confidence: 99%

Stabilization of Complementarity Systems via Contact-Aware Controllers

Aydinoglu¹,

Sieg²,

Preciado³

et al. 2020

Preprint

View full text Add to dashboard Cite

“…where k 1 = k 2 = 10 are stiffness parameters of the soft walls, g = 9.81 is the gravitational acceleration, m p = 0.5 is the mass of the pole, m c = 1 is the mass of the cart, l = 0.5 is the length of the pole, and d = 0.1 represents where the walls are. For this model, we solve the feasibility problem (13) and find a controller of the form u(x, λ) = Kx + Lλ that regulates the model to the origin. The algorithm succeeded in finding a feasible controller in 0.72 seconds.…”

Section: A Cart-pole With Soft Wallsmentioning

confidence: 99%

“…[11], [12]). Other recent methods incorporate tactile sensors within deep learning frameworks, though offer no guarantees on performance or stability [13], [14].…”

Section: Introductionmentioning

confidence: 99%

Contact-Aware Controller Design for Complementarity Systems

Aydinoglu

Preciado

Posa

2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

While many robotic tasks, like manipulation and locomotion, are fundamentally based in making and breaking contact with the environment, state-of-the-art control policies struggle to deal with the hybrid nature of multi-contact motion. Such controllers often rely heavily upon heuristics or, due to the combinatoric structure in the dynamics, are unsuitable for real-time control. Principled deployment of tactile sensors offers a promising mechanism for stable and robust control, but modern approaches often use this data in an ad hoc manner, for instance to guide guarded moves. In this work, by exploiting the complementarity structure of contact dynamics, we propose a control framework which can close the loop on rich, tactile sensors. Critically, this framework is non-combinatoric, enabling optimization algorithms to automatically synthesize provably stable control policies. We demonstrate this approach on three different underactuated, multi-contact robotics problems.

show abstract

Leveraging Contact Forces for Learning to Grasp

Cited by 2 publications

References 0 publications

Stabilization of Complementarity Systems via Contact-Aware Controllers

Stabilization of Complementarity Systems via Contact-Aware Controllers

Contact-Aware Controller Design for Complementarity Systems

Contact Info

Product

Resources

About