Robot Control Code Generation by Task Demonstration in a Dynamic Environment

Research output: Non-textual formWeb publication/site

7 Citations (Scopus)

Abstract

Generally within mobile robotics, the most dominant relationship to consider when implementing robot control code is the one between the robot’s sensors and its motors. When implementing such a relationship, efficiency and reliability are of crucial importance. The latter aspects often prove challenging due to the complex interaction between a robot and the environment in which it exists, frequently resulting in a time consuming iterative process where control code is redeveloped and tested many times before obtaining an optimal controller. In this paper, we address this challenge by implementing an alternative approach to control code generation, which first identifies the desired robot behaviour and represents the sensor-motor task algorithmically through system identification using the NARMAX modelling methodology. The control code is generated by task demonstration, where the sensory perception and velocities are logged and the relationship that exists between them is then modelled using system identification. This approach produces transparent control code through non-linear polynomial equations that can be mathematically analysed to obtain formal statements regarding specific inputs/outputs. We demonstrate this approach to control code generation and analyse its performance in dynamic environments.
LanguageEnglish
PublisherElsevier
DOIs
Publication statusPublished - 23 Aug 2012

Fingerprint

Demonstrations
Robots
Identification (control systems)
Sensors
Process control
Robotics
Code generation
Polynomials
Controllers

Cite this

@misc{840186b517e043dfb3100f534dd430de,
title = "Robot Control Code Generation by Task Demonstration in a Dynamic Environment",
abstract = "Generally within mobile robotics, the most dominant relationship to consider when implementing robot control code is the one between the robot’s sensors and its motors. When implementing such a relationship, efficiency and reliability are of crucial importance. The latter aspects often prove challenging due to the complex interaction between a robot and the environment in which it exists, frequently resulting in a time consuming iterative process where control code is redeveloped and tested many times before obtaining an optimal controller. In this paper, we address this challenge by implementing an alternative approach to control code generation, which first identifies the desired robot behaviour and represents the sensor-motor task algorithmically through system identification using the NARMAX modelling methodology. The control code is generated by task demonstration, where the sensory perception and velocities are logged and the relationship that exists between them is then modelled using system identification. This approach produces transparent control code through non-linear polynomial equations that can be mathematically analysed to obtain formal statements regarding specific inputs/outputs. We demonstrate this approach to control code generation and analyse its performance in dynamic environments.",
author = "Bryan Gardiner and SA Coleman and TM McGinnity and H He",
year = "2012",
month = "8",
day = "23",
doi = "10.1016/j.robot.2012.07.023",
language = "English",
publisher = "Elsevier",
address = "Netherlands",

}

TY - ADVS

T1 - Robot Control Code Generation by Task Demonstration in a Dynamic Environment

AU - Gardiner, Bryan

AU - Coleman, SA

AU - McGinnity, TM

AU - He, H

PY - 2012/8/23

Y1 - 2012/8/23

N2 - Generally within mobile robotics, the most dominant relationship to consider when implementing robot control code is the one between the robot’s sensors and its motors. When implementing such a relationship, efficiency and reliability are of crucial importance. The latter aspects often prove challenging due to the complex interaction between a robot and the environment in which it exists, frequently resulting in a time consuming iterative process where control code is redeveloped and tested many times before obtaining an optimal controller. In this paper, we address this challenge by implementing an alternative approach to control code generation, which first identifies the desired robot behaviour and represents the sensor-motor task algorithmically through system identification using the NARMAX modelling methodology. The control code is generated by task demonstration, where the sensory perception and velocities are logged and the relationship that exists between them is then modelled using system identification. This approach produces transparent control code through non-linear polynomial equations that can be mathematically analysed to obtain formal statements regarding specific inputs/outputs. We demonstrate this approach to control code generation and analyse its performance in dynamic environments.

AB - Generally within mobile robotics, the most dominant relationship to consider when implementing robot control code is the one between the robot’s sensors and its motors. When implementing such a relationship, efficiency and reliability are of crucial importance. The latter aspects often prove challenging due to the complex interaction between a robot and the environment in which it exists, frequently resulting in a time consuming iterative process where control code is redeveloped and tested many times before obtaining an optimal controller. In this paper, we address this challenge by implementing an alternative approach to control code generation, which first identifies the desired robot behaviour and represents the sensor-motor task algorithmically through system identification using the NARMAX modelling methodology. The control code is generated by task demonstration, where the sensory perception and velocities are logged and the relationship that exists between them is then modelled using system identification. This approach produces transparent control code through non-linear polynomial equations that can be mathematically analysed to obtain formal statements regarding specific inputs/outputs. We demonstrate this approach to control code generation and analyse its performance in dynamic environments.

U2 - 10.1016/j.robot.2012.07.023

DO - 10.1016/j.robot.2012.07.023

M3 - Web publication/site

PB - Elsevier

ER -