TY - GEN
T1 - Simulating a Multi-Agent UAV System Coordinated by State Machines Using Godot
AU - Howard, Leo
AU - Lin, Fuhua
AU - Leung, Henry
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Unmanned aerial vehicles have advanced quickly and are now a dominant force in several domains, including military, security, and even logistics. 3D simulations with virtual environments are cost effective tools for assessing feasibility and costs of developing new systems or upgrading current systems. This is especially true for multiagent or swarm systems that require many agents working in concert. The total costs can be much higher when only physical prototypes are used for testing and analysis. There are existing 3D libraries for aiding with UAV simulations, but they're too specialized; it is difficult to create highly customized virtual environments with these packages. A simulation development framework needs to be capable and stable for accurate simulations yet generalized enough that any types of simulations can be developed without exhausting efforts. This paper explores the feasibility and ease of using Godot, an open-source game engine, to develop a simulation that can assess the function of a multiagent UAV system. It is found that the Godot engine is a powerful tool that is both accurate and generalized enough to develop highly customizable UAV simulation. This allows verification and testing of UAV designs and control algorithms in highly customizable and flexible virtual environments, using a free open-sourced tool.
AB - Unmanned aerial vehicles have advanced quickly and are now a dominant force in several domains, including military, security, and even logistics. 3D simulations with virtual environments are cost effective tools for assessing feasibility and costs of developing new systems or upgrading current systems. This is especially true for multiagent or swarm systems that require many agents working in concert. The total costs can be much higher when only physical prototypes are used for testing and analysis. There are existing 3D libraries for aiding with UAV simulations, but they're too specialized; it is difficult to create highly customized virtual environments with these packages. A simulation development framework needs to be capable and stable for accurate simulations yet generalized enough that any types of simulations can be developed without exhausting efforts. This paper explores the feasibility and ease of using Godot, an open-source game engine, to develop a simulation that can assess the function of a multiagent UAV system. It is found that the Godot engine is a powerful tool that is both accurate and generalized enough to develop highly customizable UAV simulation. This allows verification and testing of UAV designs and control algorithms in highly customizable and flexible virtual environments, using a free open-sourced tool.
KW - Multiagent Systems
KW - Unmanned Aerial Vehicles
KW - component; 3D Virtual Environments
UR - https://www.scopus.com/pages/publications/105002242435
U2 - 10.1109/SWC62898.2024.00346
DO - 10.1109/SWC62898.2024.00346
M3 - Published Conference contribution
AN - SCOPUS:105002242435
T3 - Proceedings - 2024 IEEE Smart World Congress, SWC 2024 - 2024 IEEE Ubiquitous Intelligence and Computing, Autonomous and Trusted Computing, Digital Twin, Metaverse, Privacy Computing and Data Security, Scalable Computing and Communications
SP - 2273
EP - 2279
BT - Proceedings - 2024 IEEE Smart World Congress, SWC 2024 - 2024 IEEE Ubiquitous Intelligence and Computing, Autonomous and Trusted Computing, Digital Twin, Metaverse, Privacy Computing and Data Security, Scalable Computing and Communications
T2 - 10th IEEE Smart World Congress, SWC 2024
Y2 - 2 December 2024 through 7 December 2024
ER -