Why does it matter?
Over the past decades the capabilities of Unmanned Aerial Vehicles (UAVs), also known as drones, have made significant progress, primarily driven by advancements in computer processing power and electronic miniaturisation. UAVs can be used in various of real-life civil application domains including real-time monitoring, search and rescue, remote sensing, delivery of goods, precision agriculture, and infrastructure inspection, due to their vital features: effortless deployment, low maintenance cost, flexibility, adaptive altitude, high-mobility and adjustable usage. In those applications, a single UAV faces some challenges due to payload, quality and quantity of sensors, and power consumption limitations. However, a team of UAVs can be used instead to overcome some of these limitations. The range of applications of unmanned aerial vehicles (UAVs) could be extended if a team of multiple UAVs are used.
The aim of this research is to develop a framework of a team of UAVs to cooperatively finding one or multiple targets in a real-world based environment with obstacles is being developed. The framework uses a Partially Observable Markov Decision Processes (POMDP) to suit the decentralised multi-agent system while considering the necessary uncertainties in UAV localization and in the environments.
Real World Impact
For examples, applications include search and rescue, remote sensing or infrastructure inspection, which can benefit from an efficient and cooperative multi-UAV system.
Other Team Members
- Conrad Sanderson (Data61)
- Zhu, Xiaolong, Vanegas Alvarez, Fernando, Gonzalez, Felipe (2020) An approach for multi-UAV system navigation and target finding in cluttered environments Proceedings of 2020 International Conference on Unmanned Aircraft Systems: ICUAS'20, 2020 International Conference on Unmanned Aircraft Systems, ICUAS 2020, pp.1113-1120.