Computer Science
MAS4D: Advanced Multi-Agent Simulations for Sustainable Urban Air Mobility
Published on - IEEE/ACS 22nd International Conference on Computer Systems and Applications (AICCSA)
The use of Unmanned Aerial Vehicle (UAVs) has increased significantly over the past decade. They can be used for healthcare delivery, particularly for time-critical deliveries such as defibrillators and organs as well as for search and rescue operations. Effectively managing this new transport system and its traffic requires addressing scientific challenges related to complex systems, necessitating an integrated approach to the various modes of transport. Combined Multi-Agent Systems (MAS) and Reinforcement Learning (RL) algorithms can provide a relevant framework for modelling and simulating the complex interactions between UAVs, static and dynamic obstacles, and existing and future urban infrastructure, and users. The aim of thesis is to develop a new environmental framework for UAV interactions in a three-dimensional (3D) environment.