kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
BVR Gym: A Reinforcement Learning Environment for Beyond-Visual-Range Air Combat
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL. Aeronautical Solutions division, SAAB Aeronautics.ORCID iD: 0000-0003-4662-441X
Tactical Control and Data Fusion division, SAAB Aeronautics.
Aeronautical Solutions division, SAAB Aeronautics.
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0002-7714-928X
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Creating new air combat tactics and discovering novel maneuvers can require numerous hours of expert pilots' time. Additionally, for each different combat scenario, the same strategies may not work since small changes in equipment performance may drastically change the air combat outcome. For this reason, we created a reinforcement learning environment to help investigate potential air combat tactics in the field of beyond-visual-range (BVR) air combat: the BVR Gym. This type of air combat is important since long-range missiles are often the first weapon to be used in aerial combat. Some existing environments provide high-fidelity simulations but are either not open source or are not adapted to the BVR air combat domain. Other environments are open source but use less accurate simulation models. Our work provides a high-fidelity environment based on the open-source flight dynamics simulator JSBSim and is adapted to the BVR air combat domain. This article describes the building blocks of the environment and some use cases. 

National Category
Computer Systems Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-360924DOI: 10.48550/arXiv.2403.17533OAI: oai:DiVA.org:kth-360924DiVA, id: diva2:1942702
Funder
Vinnova, 2017-04875
Note

QC 20250307

Available from: 2025-03-06 Created: 2025-03-06 Last updated: 2025-03-07Bibliographically approved
In thesis
1. Data-Driven Methods for Enhanced Situation Awareness in Beyond Visual Range Air Combat
Open this publication in new window or tab >>Data-Driven Methods for Enhanced Situation Awareness in Beyond Visual Range Air Combat
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Pilots must be aware of their surroundings and environment to outperform the enemy fleet in air combat. Situation Awareness (SA) is vital. Pilots with a superior SA, compared to that of the enemy are more likely to act correctly and more quickly, which in turn increases their chances of outperforming the enemy fleet. For this reason, SA plays a significant role in the battlefield, and the techniques that provide pilots with SA must evolve with the ever-changing battlefield as air-to-air missiles' effective range increases and their performance improves.

We introduce our work in the SA domain for Beyond Visual Range (BVR) air combat. First, we describe the environment in which BVR air combat unfolds, followed by the research challenges where we address developing machine learning-driven tactics for BVR combat to optimize engagement strategies in complex and uncertain environments. Finally, we present our research results and explain how our approach can be applied to engagements with an arbitrary number of enemies and friendly units while noting that our approach should benefit both manned and unmanned aerial vehicles.

Abstract [sv]

För stridspiloter är det väldigt viktigt att vara medvetna om sin omgivning, vad gäller position och status hos både fiender och egna styrkor. Denna situationsmedvetenhet(eng. Situation Awareness, SA) är avgörande för att piloterna skall kunna agera snabbt och korrekt, och därmed vinna striden.SA är således mycket viktigt, och metoder som förbättrar SA utvecklas därför ständigt, parallellt med övrig utveckling av både materiel och taktik.

I denna avhandling presenteras vårt arbete inom SA för luftstrider där fienden befinner sig på långa avstånd (Beyond Visual Range, BVR). Först beskrivs  forskningsutmaningar med speciellt fokus på maskininlärningsdriven taktik för BVR-strider. Sedan presenterar vi våra forskningsresultat och förklarar hur de kan tillämpas i situationer med både bemannade och obemannade flygfarkoster, och med olika antal enheter på respektive sida.

Place, publisher, year, edition, pages
KTH Royal Institute of Technology, 2025. p. 66
Series
TRITA-EECS-AVL ; 2025:27
Keywords
Cooperative control, Optimization, Control Barrier Functions, Reinforcement Learning, Machine Learning, Beyond Visual Range Air Combat, Situation Awareness
National Category
Computer and Information Sciences
Research subject
Aerospace Engineering; Computer Science
Identifiers
urn:nbn:se:kth:diva-360931 (URN)978-91-8106-213-7 (ISBN)
Public defence
2025-03-28, Kollegiesalen, Brinellvägen 6, Stockholm, 10:54 (English)
Opponent
Supervisors
Funder
Vinnova, 2017-04875
Note

QC 20250306

Available from: 2025-03-06 Created: 2025-03-06 Last updated: 2025-03-17Bibliographically approved

Open Access in DiVA

fulltext(4169 kB)92 downloads
File information
File name FULLTEXT01.pdfFile size 4169 kBChecksum SHA-512
11ae9ee7a32afb8858856b58612c6b554fe700c88f16db5fe8735b3d913a03ccb57585650c920b12279502752771dfd6703d5bf0f26e26d1501e2e1a6d70ae20
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Scukins, EdvardsÖgren, Petter

Search in DiVA

By author/editor
Scukins, EdvardsÖgren, Petter
By organisation
Robotics, Perception and Learning, RPL
Computer SystemsOther Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 92 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 295 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf