1. Home page
  2. Search

Optimal Motion Planning in Constrained Workspaces using Reinforcement Learning

This item is provided by the institution :
National Technical University of Athens   

Repository :
Digital Library of National Technical University of Athens | Dspace@NTUA   

see the original item page
in the repository's web site and access all digital files if the item*



Optimal Motion Planning in Constrained Workspaces using Reinforcement Learning (EN)
Ρουσσέας, Παναγιώτης (EL)
Παπαδόπουλος, Ευάγγελος (EL)
Κυριακόπουλος, Κώστας (EL)
Γεώργιος, Βοσνιάκος (EL)
ntua (EN)
Bachelorthesis (EN)
2020-07-23
2020-10-05T16:20:42Z
In this work, a novel solution to the optimal motion planning problem is proposed, through a continuous, deterministic and provably correct approach, with guaranteed safety and which is based on a parametrized Arti cial Potential Field (APF). In particular, Reinforcement Learning (RL) is applied to adjust appropriately the parameters of the underlying potential eld towards minimizing the Hamilton-Jacobi-Bellman (HJB) error. The proposed method, outperforms consistently a Rapidly-exploring Random Trees (RRT*) method and consists a fertile advancement in the optimal motion planning problem. Finally this work gives rise to a new outlook on solutions for the aforementioned problem. (EN)
Αυτόματος έλεγχος (EL)
Ενισχυτική μάθηση (EL)
Τεχνητή νοημοσύνη (EL)
Ρομποτική (EL)
Σχεδιασμός πορείας (EL)
Motion planning (EN)
Reinforcement learning (EN)
Control systems (EN)
Artificial intelligence (EN)
Robotics (EN)
English
Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Μηχανολόγων Μηχανικών. Τομέας Μηχανολογικών Κατασκευών και Αυτομάτου Ελέγχου (EL)
Εργαστήριο Αυτομάτου Ελέγχου (EL)
Αναφορά Δημιουργού 3.0 Ελλάδα
http://creativecommons.org/licenses/by/3.0/gr/
National Technical University of Athens
Digital Library of National Technical University of Athens | Dspace@NTUA
Κεντρική Βιβλιοθήκη Ε.Μ.Π.
Ιδρυματικό Αποθετήριο
Διπλωματικές Εργασίες



*Institutions are responsible for keeping their URLs functional (digital file, item page in repository site)