Teaching a Vehicle to Autonomously Drift: A Data-based Approach Using Neural Networks

Manuel Acosta, Stratis Kanarachos

Research output: Contribution to journalArticle

4 Citations (Scopus)
17 Downloads (Pure)

Abstract

This paper presents a novel approach to teach a vehicle how to drift, in a
similar manner that professional drivers do. Specically, a hybrid structure
formed by a Model Predictive Controller and feedforward Neural Networks
is employed for this purpose. The novelty of this work lies in a) the adoption
of a data-based approach to achieve autonomous drifting along a wide range
of road radii and body slip angles, and b) in the implementation of a road
terrain classier to adjust the system actuation depending on the current
friction characteristics. The presented drift control system is implemented
in a multi-actuated ground vehicle equipped with active front steering and
in-wheel electric motors and trained to drift by a real test driver using a
driver-in-the-loop setup. Its performance is veried in the simulation environment
IPG-CarMaker through dierent open loop and path following
drifting manoeuvres.
Original languageEnglish
Pages (from-to)12-28
Number of pages17
JournalKnowledge-Based Systems
Volume153
Early online date11 Apr 2018
DOIs
Publication statusPublished - 1 Aug 2018

Fingerprint

Ground vehicles
Electric motors
Wheels
Teaching
Neural networks
Control systems
Controllers

Bibliographical note

NOTICE: this is the author’s version of a work that was accepted for publication in Knowledge-Based Systems. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Knowledge-Based Systems, [153], (2018) DOI: 10.1016/j.knosys.2018.04.015

Keywords

  • Neural Networks
  • Autonomous Drift Control
  • Autonomous Vehicles
  • Multi-Actuated Ground Vehicles
  • Model Predictive Control

Cite this

Teaching a Vehicle to Autonomously Drift: A Data-based Approach Using Neural Networks. / Acosta, Manuel; Kanarachos, Stratis.

In: Knowledge-Based Systems, Vol. 153, 01.08.2018, p. 12-28.

Research output: Contribution to journalArticle

@article{74dd8ed74b954daaaf9afa6c822b49fd,
title = "Teaching a Vehicle to Autonomously Drift: A Data-based Approach Using Neural Networks",
abstract = "This paper presents a novel approach to teach a vehicle how to drift, in asimilar manner that professional drivers do. Specically, a hybrid structureformed by a Model Predictive Controller and feedforward Neural Networksis employed for this purpose. The novelty of this work lies in a) the adoptionof a data-based approach to achieve autonomous drifting along a wide rangeof road radii and body slip angles, and b) in the implementation of a roadterrain classier to adjust the system actuation depending on the currentfriction characteristics. The presented drift control system is implementedin a multi-actuated ground vehicle equipped with active front steering andin-wheel electric motors and trained to drift by a real test driver using adriver-in-the-loop setup. Its performance is veried in the simulation environmentIPG-CarMaker through dierent open loop and path followingdrifting manoeuvres.",
keywords = "Neural Networks, Autonomous Drift Control, Autonomous Vehicles, Multi-Actuated Ground Vehicles, Model Predictive Control",
author = "Manuel Acosta and Stratis Kanarachos",
note = "NOTICE: this is the author’s version of a work that was accepted for publication in Knowledge-Based Systems. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Knowledge-Based Systems, [153], (2018) DOI: 10.1016/j.knosys.2018.04.015",
year = "2018",
month = "8",
day = "1",
doi = "10.1016/j.knosys.2018.04.015",
language = "English",
volume = "153",
pages = "12--28",
journal = "Knowledge-Based Systems",
issn = "0950-7051",
publisher = "Elsevier",

}

TY - JOUR

T1 - Teaching a Vehicle to Autonomously Drift: A Data-based Approach Using Neural Networks

AU - Acosta, Manuel

AU - Kanarachos, Stratis

N1 - NOTICE: this is the author’s version of a work that was accepted for publication in Knowledge-Based Systems. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Knowledge-Based Systems, [153], (2018) DOI: 10.1016/j.knosys.2018.04.015

PY - 2018/8/1

Y1 - 2018/8/1

N2 - This paper presents a novel approach to teach a vehicle how to drift, in asimilar manner that professional drivers do. Specically, a hybrid structureformed by a Model Predictive Controller and feedforward Neural Networksis employed for this purpose. The novelty of this work lies in a) the adoptionof a data-based approach to achieve autonomous drifting along a wide rangeof road radii and body slip angles, and b) in the implementation of a roadterrain classier to adjust the system actuation depending on the currentfriction characteristics. The presented drift control system is implementedin a multi-actuated ground vehicle equipped with active front steering andin-wheel electric motors and trained to drift by a real test driver using adriver-in-the-loop setup. Its performance is veried in the simulation environmentIPG-CarMaker through dierent open loop and path followingdrifting manoeuvres.

AB - This paper presents a novel approach to teach a vehicle how to drift, in asimilar manner that professional drivers do. Specically, a hybrid structureformed by a Model Predictive Controller and feedforward Neural Networksis employed for this purpose. The novelty of this work lies in a) the adoptionof a data-based approach to achieve autonomous drifting along a wide rangeof road radii and body slip angles, and b) in the implementation of a roadterrain classier to adjust the system actuation depending on the currentfriction characteristics. The presented drift control system is implementedin a multi-actuated ground vehicle equipped with active front steering andin-wheel electric motors and trained to drift by a real test driver using adriver-in-the-loop setup. Its performance is veried in the simulation environmentIPG-CarMaker through dierent open loop and path followingdrifting manoeuvres.

KW - Neural Networks

KW - Autonomous Drift Control

KW - Autonomous Vehicles

KW - Multi-Actuated Ground Vehicles

KW - Model Predictive Control

U2 - 10.1016/j.knosys.2018.04.015

DO - 10.1016/j.knosys.2018.04.015

M3 - Article

VL - 153

SP - 12

EP - 28

JO - Knowledge-Based Systems

JF - Knowledge-Based Systems

SN - 0950-7051

ER -