Back to search

Virginia Tech

Deploying Reinforcement Learning in the Real World: A Case Study on Apptronik Apollo

Abstract

dc:description.abstract

Humanoid robots pose significant challenges for modeling and control: their many degrees of freedom and complex contact dynamics force traditional controllers to rely on simplified physical models that limit agility and robustness. Deep reinforcement learning offers an alternative: policies represented by deep neural networks can discover dynamic behaviors directly through millions of trials in a high-fidelity simulator, eliminating the need for simplified models. This thesis develops and validates a sim-to-real pipeline for deploying such policies on a humanoid robot, demonstrating reliable transfer of dynamic balancing and teleoperation skills from simulation to the Apptronik Apollo robot in the real world.

Degree

thesis:*
Name thesis:degree_name
Master of Science
Level thesis:degree_level
masters
Discipline thesis:degree_discipline
Electrical Engineering
Department dc:contributor.department
Electrical Engineering
Grantor dc:publisher
Virginia Tech
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Welch, Stephen Brian
Chair dc:contributor.committeechair
  • Stilwell, Daniel J.
Committee members dc:contributor.committeemember
  • Williams, Ryan K.
  • Leonessa, Alexander

Subjects

dc:subject × 3

Rights

dc:rights
Statement dc:rights
  • In Copyright
Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Other
vt_gsexam:44175

Chain of custody

source
Harvested from
Virginia Tech
Base URL
vtechworks.lib.vt.edu/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Welch, Stephen Brian. Deploying Reinforcement Learning in the Real World: A Case Study on Apptronik Apollo. masters thesis, Virginia Tech, 2025. https://hdl.handle.net/10919/135528