Learning motion primitive goals for robust manipulation

By Freek Stulp, Evangelos A. Theodorou, Mrinal Kalakrishnan, Peter Pastor, Ludovic Righetti, Stefan Schaal. 2011

49 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Cognitive Neuroscience, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Gerhard Neumann
2. Jan Peters
3. Martin A. Riedmiller
4. Tamim Asfour
5. Danica Kragic
6. Thomas Lampe
7. Jeannette Bohg
8. Sergey Levine

In Cognitive Neuroscience

1. Gerhard Neumann
2. Takayuki Osa
3. Jan Peters
4. Riad Akrour
5. Tamim Asfour
6. Danica Kragic
7. Jeannette Bohg
8. Thomas Lampe

In Control and Systems Engineering

1. Sergey Levine
2. Gerhard Neumann
3. Jan Peters
4. Pieter Abbeel
5. Thomas Lampe
6. Martin A. Riedmiller
7. Jeannette Bohg
8. Danica Kragic

Citations in the corpus

• Active learning of inverse models with intrinsically motivated goal exploration in robots (2013, neutral): “For the optimization part, algorithms such as natural actor-critic architectures in model based reinforcement learning [98], algorithms of convex optimization [100], algorithms of stochastic optimization like CMA (e.g. [99]), or path-integral methods (e.g. [101,84]).”
• Data-Driven Grasp Synthesis—A Survey (2013, methodology): “Similar initialization of an object specific grasping policy is used in Pastor et al. [25] and Stulp et al. [50].”
• Sample and Feedback Efficient Hierarchical Reinforcement Learning from Human Preferences (2018, methodology): “[24] employ RL to learn the goal and shape parameters of a movement primitive.”
• Acquiring visual servoing reaching and grasping skills using neural reinforcement learning (2013, methodology): “generation of trajectories using only a generic initializa tion [6], [23].”