Learning an internal representation of the end-effector configuration space

TL;DR

This paper introduces a method to learn an internal representation of a robot's end-effector configuration space solely from raw sensor data, without prior structural or sensor information, enabling control based on this learned model.

Contribution

It presents a novel approach to estimate a robot's end-effector configuration internally from unstructured data without relying on known kinematic models or sensor specifics.

Findings

Successfully generates internal end-effector representations from raw data

Enables robot control using learned internal representations

Operates with minimal assumptions about robot structure

Abstract

Current machine learning techniques proposed to automatically discover a robot kinematics usually rely on a priori information about the robot's structure, sensors properties or end-effector position. This paper proposes a method to estimate a certain aspect of the forward kinematics model with no such information. An internal representation of the end-effector configuration is generated from unstructured proprioceptive and exteroceptive data flow under very limited assumptions. A mapping from the proprioceptive space to this representational space can then be used to control the robot.