Model Based Reinforcement Learning for Autonomous Headland Turns

Institut
Lehrstuhl für Agrarmechatronik (WZW)
Typ
Semesterarbeit / Masterarbeit /
Inhalt
experimentell / theoretisch /  
Beschreibung

Minimizing non-working time is a major goal for efficient automated agriculture. Much of this time is
spent turning in the headland. We want to investigate, if Model Based RL can be used to learn the
real vehicle’s model and use this knowledge to quickly optimize the turning behaviour.

Voraussetzungen
  • A Bachelor’s level university degree
  • Previous Experience with Reinforcement Learning and Mobile Robotics
  • An Interest in Agricultural Machinery
Möglicher Beginn
sofort
Kontakt
Lukas Pindl
lukas.pindltum.de
Ausschreibung