Latent spaces for dynamic movement primitives

Bitzer, Sebastian; Vijayakumar, Sethu

doi:10.1109/ichr.2009.5379530

Cited by 41 publications

(31 citation statements)

References 23 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of the work presented here is to study and model the spatial dimensionality and temporal context of human hand actions to i) provide representations for programming grasping actions in robots, and ii) use these for designing new robotic and prosthetic hands. In robotics, it has been argued that continuous motion mapping from human to robot requires suitable spatio-temporal representations of human and robot motions, [1], [2]. However, most of the work on grasp mapping is based on handdesigned grasping taxonomies considering a discrete set of hand postures, [3], [4].…”

Section: Introductionmentioning

confidence: 99%

Spatio-temporal modeling of grasping actions

Romero¹,

Feix

Kragić³

2010

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

Abstract-Understanding the spatial dimensionality and temporal context of human hand actions can provide representations for programming grasping actions in robots and inspire design of new robotic and prosthetic hands. The natural representation of human hand motion has high dimensionality. For specific activities such as handling and grasping of objects, the commonly observed hand motions lie on a lower-dimensional non-linear manifold in hand posture space. Although full body human motion is well studied within Computer Vision and Biomechanics, there is very little work on the analysis of hand motion with nonlinear dimensionality reduction techniques. In this paper we use Gaussian Process Latent Variable Models (GPLVMs) to model the lower dimensional manifold of human hand motions during object grasping. We show how the technique can be used to embed high-dimensional grasping actions in a lower-dimensional space suitable for modeling, recognition and mapping.

show abstract

Section: Introductionmentioning

confidence: 99%

Spatio-temporal modeling of grasping actions

Romero¹,

Feix

Kragić³

2010

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

show abstract

“…In [2], a latent manifold in joint space is computed using Gaussian process from sample configurations produced by an expert. This manifold is, however, defined by samples from a valid trajectory in joint space and it does not capture state of the environment directly.…”

Section: A Previous Workmentioning

confidence: 99%

Hierarchical Motion Planning in Topological Representations

Zarubin

Ivan²,

Toussaint

et al. 2012

Robotics: Science and Systems VIII

Self Cite

View full text Add to dashboard Cite

Abstract-Motion can be described in alternative representations, including joint configuration or end-effector spaces, but also more complex topological representations that imply a change of Voronoi bias, metric or topology of the motion space. Certain types of robot interaction problems, e.g. wrapping around an object, can suitably be described by so-called writhe and interaction mesh representations. However, considering motion synthesis solely in topological spaces is insufficient since it does not cater for additional tasks and constraints in other representations. In this paper we propose methods to combine and exploit different representations for motion synthesis, with specific emphasis on generalization of motion to novel situations. Our approach is formulated in the framework of optimal control as an approximate inference problem, which allows for a direct extension of the graphical model to incorporate multiple representations. Motion generalization is similarly performed by projecting motion from topological to joint configuration space. We demonstrate the benefits of our methods on problems where direct path finding in joint configuration space is extremely hard whereas local optimal control exploiting a representation with different topology can efficiently find optimal trajectories. Further, we illustrate the successful online motion generalization to dynamic environments on challenging, real world problems.

show abstract

“…However, all the available data could be used more efficiently if a principled way to integrate this data was designed. However, it is well known that the average of solutions to a problem is often not itself a solution, making this problem non-trivial, though some approaches have been explored [Bitzer and Vijayakumar, 2009;Coates et al, 2008]. Furthermore, by using DMPs, we are performing somewhat "blind" replay of skills.…”

Section: Future Workmentioning

confidence: 99%

Learning grounded finite-state representations from unstructured demonstrations

Niekum

Osentoski

Konidaris³

et al. 2014

The International Journal of Robotics Research

178

160

View full text Add to dashboard Cite

Robots exhibit flexible behavior largely in proportion to their degree of knowledge about the world. Such knowledge is often meticulously hand-coded for a narrow class of tasks, limiting the scope of possible robot competencies. Thus, the primary limiting factor of robot capabilities is often not the physical attributes of the robot, but the limited time and skill of expert programmers. One way to deal with the vast number of situations and environments that robots face outside the laboratory is to provide users with simple methods for programming robots that do not require the skill of an expert.For this reason, learning from demonstration (LfD) has become a popular alternative to traditional robot programming methods, aiming to provide a natural mechanism for quickly teaching robots. By simply showing a robot how to perform a task, users can easily demonstrate new tasks as needed, without any special knowledge about the robot. Unfortunately, LfD often yields little knowledge about the world, and thus lacks robust generalization capabilities, especially for complex, multi-step tasks.We present a series of algorithms that draw from recent advances in Bayesian nonparametric statistics and control theory to automatically detect and leverage repeated structure at multiple levels of abstraction in demonstration data. The discovery of repeated structure provides critical insights into task invariants, features of importance, high-level task structure, and appropriate skills for the task. This culminates in the discovery of a finite-state representation of the task, comprised of grounded skills that are flexible and reusable, providing robust generalization and transfer in complex, multistep robotic tasks. These algorithms are tested and evaluated using a PR2 mobile manipulator, showing success on several complex real-world tasks, such as furniture assembly.

show abstract

Latent spaces for dynamic movement primitives

Cited by 41 publications

References 23 publications

Spatio-temporal modeling of grasping actions

Spatio-temporal modeling of grasping actions

Hierarchical Motion Planning in Topological Representations

Learning grounded finite-state representations from unstructured demonstrations

Contact Info

Product

Resources

About