Towards Automated Code Generation for Autonomous Mobile Robots

D Kerr, U Nehmzow, S.A. Billings

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

With the expected growth in mobile robotics the demand for expertise to develop robot control code will also increase. As end-users cannot be expected to develop this control code themselves, a more elegant solution would be to allow the end-users to teach the robot by demonstrating the task. In this paper we show how route learning tasks may betranslated directly into robot control code simply by observing the task. We show how automated code generation may be facilitated through system identification which algorithmically and automatically transfers human behaviour into control code, using trans-parent mathematical functions. We provide two route learning examples where a mobile robot automatically obtains control code simply by observing human behaviour, identifying it using system identication, and copying the behaviour.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages55-60
Number of pages5
DOIs
Publication statusPublished - 2010
EventThe Third Conference on Artificial General Intelligence - Lugano. Switzerland
Duration: 1 Jan 2010 → …

Conference

ConferenceThe Third Conference on Artificial General Intelligence
Period1/01/10 → …

Fingerprint

Mobile robots
Robots
Copying
Identification (control systems)
Robotics
Code generation

Cite this

Kerr, D., Nehmzow, U., & Billings, S. A. (2010). Towards Automated Code Generation for Autonomous Mobile Robots. In Unknown Host Publication (pp. 55-60) https://doi.org/10.2991/agi.2010.37
Kerr, D ; Nehmzow, U ; Billings, S.A. / Towards Automated Code Generation for Autonomous Mobile Robots. Unknown Host Publication. 2010. pp. 55-60
@inproceedings{070ed5c518344093b69b16f45b237654,
title = "Towards Automated Code Generation for Autonomous Mobile Robots",
abstract = "With the expected growth in mobile robotics the demand for expertise to develop robot control code will also increase. As end-users cannot be expected to develop this control code themselves, a more elegant solution would be to allow the end-users to teach the robot by demonstrating the task. In this paper we show how route learning tasks may betranslated directly into robot control code simply by observing the task. We show how automated code generation may be facilitated through system identification which algorithmically and automatically transfers human behaviour into control code, using trans-parent mathematical functions. We provide two route learning examples where a mobile robot automatically obtains control code simply by observing human behaviour, identifying it using system identication, and copying the behaviour.",
author = "D Kerr and U Nehmzow and S.A. Billings",
year = "2010",
doi = "10.2991/agi.2010.37",
language = "English",
isbn = "978-90-78677-36-9",
pages = "55--60",
booktitle = "Unknown Host Publication",

}

Kerr, D, Nehmzow, U & Billings, SA 2010, Towards Automated Code Generation for Autonomous Mobile Robots. in Unknown Host Publication. pp. 55-60, The Third Conference on Artificial General Intelligence, 1/01/10. https://doi.org/10.2991/agi.2010.37

Towards Automated Code Generation for Autonomous Mobile Robots. / Kerr, D; Nehmzow, U; Billings, S.A.

Unknown Host Publication. 2010. p. 55-60.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Towards Automated Code Generation for Autonomous Mobile Robots

AU - Kerr, D

AU - Nehmzow, U

AU - Billings, S.A.

PY - 2010

Y1 - 2010

N2 - With the expected growth in mobile robotics the demand for expertise to develop robot control code will also increase. As end-users cannot be expected to develop this control code themselves, a more elegant solution would be to allow the end-users to teach the robot by demonstrating the task. In this paper we show how route learning tasks may betranslated directly into robot control code simply by observing the task. We show how automated code generation may be facilitated through system identification which algorithmically and automatically transfers human behaviour into control code, using trans-parent mathematical functions. We provide two route learning examples where a mobile robot automatically obtains control code simply by observing human behaviour, identifying it using system identication, and copying the behaviour.

AB - With the expected growth in mobile robotics the demand for expertise to develop robot control code will also increase. As end-users cannot be expected to develop this control code themselves, a more elegant solution would be to allow the end-users to teach the robot by demonstrating the task. In this paper we show how route learning tasks may betranslated directly into robot control code simply by observing the task. We show how automated code generation may be facilitated through system identification which algorithmically and automatically transfers human behaviour into control code, using trans-parent mathematical functions. We provide two route learning examples where a mobile robot automatically obtains control code simply by observing human behaviour, identifying it using system identication, and copying the behaviour.

U2 - 10.2991/agi.2010.37

DO - 10.2991/agi.2010.37

M3 - Conference contribution

SN - 978-90-78677-36-9

SP - 55

EP - 60

BT - Unknown Host Publication

ER -

Kerr D, Nehmzow U, Billings SA. Towards Automated Code Generation for Autonomous Mobile Robots. In Unknown Host Publication. 2010. p. 55-60 https://doi.org/10.2991/agi.2010.37