Reaching the limit in autonomous racing: Optimal control versus reinforcement learning

A central question in robotics is how to design a control system for an agile mobile robot. This paper studies this question systematically, focusing on a challenging setting: autonomous drone racing. We show that a neural network controller trained with reinforcement learning (RL) outperformed opti...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Science robotics 2023-09, Vol.8 (82), p.eadg1462-eadg1462
Hauptverfasser:	Song, Yunlong, Romero, Angel, Müller, Matthias, Koltun, Vladlen, Scaramuzza, Davide
Format:	Artikel
Sprache:	eng
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!