Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation

By Dov Katz, Yuri Pyuro, Oliver Brock. 2008

91 citations recorded by the source corpus.

We introduce a learning-based approach to manipulation in unstructured environments. This approach permits autonomous acquisition of manipulation expertise from interactions with the environment. The resulting expertise enables a robot to perform effective manipulation based on partial state information. The manipulation expertise is represented in a relational state representation and learned using relational reinforcement learning. The relational representation renders learning tractable by collapsing a large number of states onto a single, relational state. The relational state representation is carefully grounded in the perceptual and interaction skills of the robot. This ensures that symbolically learned knowledge remains meaningful in the physical world. We experimentally validate the proposed learning approach on the task of manipulating an articulated object to obtain a model of its kinematic structure. Our experiments demonstrate that the manipulation expertise acquired by the robot leads to substantial performance improvements. These improvements are maintained when experience is applied to previously unseen objects.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Control and Systems Engineering

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, background): “This representation may be preferable for highly geometric tasks (which frequently appear in robotics) and has been employed to learn to navigate buildings with a real robot in a supervised setting (Cocora et al., 2006) and to manipulate articulated objects in simulation (Katz et al., 2008).”
Learning robot in-hand manipulation with tactile features (2015, background): “In previous work, reinforcement learning (RL) has been used to learn how to manipulate objects on a symbolic level [29].”
Solving Relational and First-Order Logical Markov Decision Processes: A Survey (2012, methodology): “Katz et al (2008) use a relational instance-based algorithm in a robotic manipulation task.”
Interactive Perception: Leveraging Action in Perception and Perception in Action (2016, background): “del Estimation Otte et al. [68] Pillai et al. [16] Hausman et al. [33] Barragan et al. [60]´ Mart´ın Mart ´ın and Brock [17] Jain and Kemp [56] Sturm et al. [15] Karayiannidis et al. [57] Katz et al. [69] Katz et al. [14] Fig. 6. Paper categorization based on application areas. Papers that address m”
Body schema acquisition through active learning (2010, methodology): “A similar approach is used in [ 11 ] to learn generic planar kinematics of tools and articulated objects, using a reinforcement learning model based on the number of articulations discovered.”