advertisement

Learning Kinematics from direct Self-Observation using Nearest-Neighbour Methods

50 %
50 %
advertisement
Information about Learning Kinematics from direct Self-Observation using Nearest-Neighbour...
Technology

Published on November 3, 2008

Author: cijat

Source: slideshare.net

Description

Commonly, the kinematic function of robotic manipulators is derived from the robot model analytically and can be represented in closed form. However, there are cases in which a model is not apriori available. We propose an approach that enables an autonomous robot to estimate the kinematic and inverse kinematic function on-the-fly directly from self-observations. As observations, we sample pairs of randomly chosen joint configurations and the resulting world positions. For approximating the kinematic and inverse kinematic function, we use instance-based learning techniques, such as Nearest Neighbour (NN) and Locally Weighted Regression (LWR). The sampled pairs not only contain information about the kinematics, but also implicitly encode the connectivity and reachability both in the configuration and world space. The robot can take advantage of this information to build roadmaps with a combined cost function. We present an analysis of our approach as well as the results obtained from experiments on a real robot and from simulation. We show in our talk, that with our approach, it becomes possible to accurately control robots with unknown kinematic models of various complexity and joint types from little data obtained through self-observation.
advertisement

Add a comment

Related presentations

Presentación que realice en el Evento Nacional de Gobierno Abierto, realizado los ...

In this presentation we will describe our experience developing with a highly dyna...

Presentation to the LITA Forum 7th November 2014 Albuquerque, NM

Un recorrido por los cambios que nos generará el wearabletech en el futuro

Um paralelo entre as novidades & mercado em Wearable Computing e Tecnologias Assis...

Microsoft finally joins the smartwatch and fitness tracker game by introducing the...

Related pages

Learning Kinematics from Direct Self-Observation Using ...

Learning Kinematics from Direct Self ... of learning kinematics of robots ... Self-Observation using Nearest-Neighbor Methods 9 ...
Read more

Learning Kinematics from Direct Self-Observation Using ...

online learning methods for ... Learning Kinematics from Direct Self-Observation using ... Our system used nearest-neighbour ...
Read more

Learning Kinematics from Direct Self-Observation Using ...

Publication » Learning Kinematics from Direct Self-Observation Using Nearest-Neighbor Methods. ... Learning Kinematics from Direct Self-Observation U...
Read more

www.science.gov

www.science.gov
Read more

Hapter2 Direct Kinematics - Documents

Hapter2 Direct Kinematics. by austinvishal. on Oct 16, 2015. Report Category: Documents. Download: 0 Comment: 0. 6. views. Comments. Description.
Read more

Dr. Juergen Sturm - Bib - Technische Universität München

... {van Soest} and Marc {de Greef} and J{"u}rgen Sturm and Arnoud Visser", title = "Autonomous Color Learning in an Artificial Environment" ...
Read more

The contents of this booklet are for information purposes only

... following which students work together in teams engaging in experiential learning in assessing ... Statics, Kinematics ... and Direct Effect ...
Read more

Raising expectations in GDA agents acting in dynamic ...

Representation learning methods, which project the original features to a new feature space, ... we propose a framework for using reinforcement learning ...
Read more