Skip to main navigation Skip to search Skip to main content

Cessna Citation X Ground Speed Control using Reinforcement Learning

  • École de technologie supérieure

Research output: Contribution to Book/Report types › Contribution to conference proceedings › peer-review

2 Citations (Scopus)

Abstract

Throughout this study, a novel approach was developed at the Laboratory of Applied Research in Active Control, Avionics, and AeroServoElasticity (LARCASE) to control the ground speed of an aircraft using Reinforcement Learning. For this purpose, a Reinforcement Learning controller using a Deep Deterministic Policy Gradient agent was developed. This study used a non-linear longitudinal ground dynamics model of the Cessna Citation X aircraft that was identified and validated using a level-D Research Aircraft Flight Simulator (RAFS). During the training process, the DDPG agent learned to control the aircraft ground speed by controlling the engine fan speed (N1P) and the brake. To validate the proposed model, several simulations were performed with different ground speed profiles to evaluate its performance under different acceleration and deceleration conditions. Overall, the results indicated that the controller maintained the desired reference speed, regardless of the acceleration or deceleration profiles.

Original languageEnglish
Title of host publicationAIAA SciTech Forum and Exposition, 2024
PublisherAmerican Institute of Aeronautics and Astronautics Inc, AIAA
ISBN (Print)9781624107115
DOIs
Publication statusPublished - 2024
Externally publishedYes
EventAIAA SciTech Forum and Exposition, 2024 - Orlando, United States
Duration: 8 Jan 2024 → 12 Jan 2024

Publication series

NameAIAA SciTech Forum and Exposition, 2024

Conference

ConferenceAIAA SciTech Forum and Exposition, 2024
Country/TerritoryUnited States
CityOrlando
Period8/01/24 → 12/01/24

ASJC Scopus subject areas

  • Aerospace Engineering

Fingerprint

Dive into the research topics of 'Cessna Citation X Ground Speed Control using Reinforcement Learning'. These topics are generated from the title and abstract of the publication. Together, they form a unique fingerprint.

Cite this