Abstract

Task-parameterized models of movements/behaviors refer to representations that can readily adapt to a set of task parameters describing the current situation encountered by the robot, such as the location of objects in its workspace. This paper starts with an overview of the task-parameterized Gaussian mixture model (TP-GMM) introduced in previous publications, and then introduces a number of extensions and ongoing challenges required to move the approach toward unconstrained environments. In particular, it discusses its generalization capability and the handling of movements with a high number of degrees of freedom. It then shows that the method is not restricted to movements in task space, but that it can also be exploited to handle constraints in joint space, including priority constraints.

Bibtex reference

@Inbook{Calinon18ISRR,
	author="Calinon, S.",
	title="Robot Learning with Task-Parameterized Generative Models",
	booktitle="Robotics Research",
	year="2018",
	volume="3",
	pages="111--126",
	editor="Bicchi, A. and Burgard, W.",
	publisher="Springer International Publishing",
	doi="10.1007/978-3-319-60916-4_7",
}

Source codes

Source codes related to this publication are available as part of PbDlib.

Go back to the list of publications