Optimal Motion Planning in Unknown Workspaces Using Integral Reinforcement Learning

Panagiotis Rousseas; Charalampos P. Bechlioulis; Kostas J. Kyriakopoulos

doi:10.1109/LRA.2022.3178788

Optimal Motion Planning in Unknown Workspaces Using Integral Reinforcement Learning

Panagiotis Rousseas, Charalampos P. Bechlioulis, Kostas J. Kyriakopoulos

Electrical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

A novel motion planning scheme for optimal navigation in unknown workspaces is proposed in this letter. Based upon the Artificial Harmonic Potential Fields (AHPFs) theory, a robust framework for provably correct (i.e., safe and globally convergent) navigation is enhanced through Integral Reinforcement Learning (IRL)1 to obtain a provably complete solution for optimal motion planning in unknown workspaces. Our method aims at bridging the gap between the control theoretic framework of mathematical rigor, with the data-driven Reinforcement Learning (RL) paradigm, while preserving the strong traits of each approach. Finally, it is compared against an RRT∗ method to asses the optimality of the final results in a multiply connected synthetic workspace.

Original language	English (US)
Pages (from-to)	6926-6933
Number of pages	8
Journal	IEEE Robotics and Automation Letters
Volume	7
Issue number	3
DOIs	https://doi.org/10.1109/LRA.2022.3178788
State	Published - Jul 1 2022

Keywords

Motion and path planning
optimization and optimal control
reinforcement learning

ASJC Scopus subject areas

Control and Systems Engineering
Biomedical Engineering
Human-Computer Interaction
Mechanical Engineering
Computer Vision and Pattern Recognition
Computer Science Applications
Control and Optimization
Artificial Intelligence

Access to Document

10.1109/LRA.2022.3178788

Cite this

@article{f8bc54a76df24dad97e77d146c46358a,

title = "Optimal Motion Planning in Unknown Workspaces Using Integral Reinforcement Learning",

abstract = "A novel motion planning scheme for optimal navigation in unknown workspaces is proposed in this letter. Based upon the Artificial Harmonic Potential Fields (AHPFs) theory, a robust framework for provably correct (i.e., safe and globally convergent) navigation is enhanced through Integral Reinforcement Learning (IRL)1 to obtain a provably complete solution for optimal motion planning in unknown workspaces. Our method aims at bridging the gap between the control theoretic framework of mathematical rigor, with the data-driven Reinforcement Learning (RL) paradigm, while preserving the strong traits of each approach. Finally, it is compared against an RRT∗ method to asses the optimality of the final results in a multiply connected synthetic workspace.",

keywords = "Motion and path planning, optimization and optimal control, reinforcement learning",

author = "Panagiotis Rousseas and Bechlioulis, {Charalampos P.} and Kyriakopoulos, {Kostas J.}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.",

year = "2022",

month = jul,

day = "1",

doi = "10.1109/LRA.2022.3178788",

language = "English (US)",

volume = "7",

pages = "6926--6933",

journal = "IEEE Robotics and Automation Letters",

issn = "2377-3766",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - Optimal Motion Planning in Unknown Workspaces Using Integral Reinforcement Learning

AU - Rousseas, Panagiotis

AU - Bechlioulis, Charalampos P.

AU - Kyriakopoulos, Kostas J.

PY - 2022/7/1

Y1 - 2022/7/1

N2 - A novel motion planning scheme for optimal navigation in unknown workspaces is proposed in this letter. Based upon the Artificial Harmonic Potential Fields (AHPFs) theory, a robust framework for provably correct (i.e., safe and globally convergent) navigation is enhanced through Integral Reinforcement Learning (IRL)1 to obtain a provably complete solution for optimal motion planning in unknown workspaces. Our method aims at bridging the gap between the control theoretic framework of mathematical rigor, with the data-driven Reinforcement Learning (RL) paradigm, while preserving the strong traits of each approach. Finally, it is compared against an RRT∗ method to asses the optimality of the final results in a multiply connected synthetic workspace.

AB - A novel motion planning scheme for optimal navigation in unknown workspaces is proposed in this letter. Based upon the Artificial Harmonic Potential Fields (AHPFs) theory, a robust framework for provably correct (i.e., safe and globally convergent) navigation is enhanced through Integral Reinforcement Learning (IRL)1 to obtain a provably complete solution for optimal motion planning in unknown workspaces. Our method aims at bridging the gap between the control theoretic framework of mathematical rigor, with the data-driven Reinforcement Learning (RL) paradigm, while preserving the strong traits of each approach. Finally, it is compared against an RRT∗ method to asses the optimality of the final results in a multiply connected synthetic workspace.

KW - Motion and path planning

KW - optimization and optimal control

KW - reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85131758863&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85131758863&partnerID=8YFLogxK

U2 - 10.1109/LRA.2022.3178788

DO - 10.1109/LRA.2022.3178788

M3 - Article

AN - SCOPUS:85131758863

SN - 2377-3766

VL - 7

SP - 6926

EP - 6933

JO - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

IS - 3

ER -

Optimal Motion Planning in Unknown Workspaces Using Integral Reinforcement Learning

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this