Saved in:
Bibliographic Details
Main Authors: Jiang, Shuo, Zhang, Jinkun, Wong, Lawson
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.18675
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910348287672320
author Jiang, Shuo
Zhang, Jinkun
Wong, Lawson
author_facet Jiang, Shuo
Zhang, Jinkun
Wong, Lawson
contents For a robot, its body structure is an a-prior knowledge when it is designed. However, when such information is not available, can a robot recognize it by itself? In this paper, we aim to grant a robot such ability to learn its body structure from exteroception and proprioception data collected from on-body sensors. By a novel machine learning method, the robot can learn a binary Heterogeneous Dependency Matrix from its sensor readings. We showed such matrix is equivalent to a Heterogeneous out-tree structure which can uniquely represent the robot body topology. We explored the properties of such matrix and the out-tree, and proposed a remedy to fix them when they are contaminated by partial observability or data noise. We ran our algorithm on 6 different robots with different body structures in simulation and 1 real robot. Our algorithm correctly recognized their body structures with only on-body sensor readings but no topology prior knowledge.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18675
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Robot Body Schema Learning from Full-body Extero/Proprioception Sensors
Jiang, Shuo
Zhang, Jinkun
Wong, Lawson
Robotics
For a robot, its body structure is an a-prior knowledge when it is designed. However, when such information is not available, can a robot recognize it by itself? In this paper, we aim to grant a robot such ability to learn its body structure from exteroception and proprioception data collected from on-body sensors. By a novel machine learning method, the robot can learn a binary Heterogeneous Dependency Matrix from its sensor readings. We showed such matrix is equivalent to a Heterogeneous out-tree structure which can uniquely represent the robot body topology. We explored the properties of such matrix and the out-tree, and proposed a remedy to fix them when they are contaminated by partial observability or data noise. We ran our algorithm on 6 different robots with different body structures in simulation and 1 real robot. Our algorithm correctly recognized their body structures with only on-body sensor readings but no topology prior knowledge.
title Robot Body Schema Learning from Full-body Extero/Proprioception Sensors
topic Robotics
url https://arxiv.org/abs/2402.18675