Page 17 - AI Ready – Analysis Towards a Standardized Readiness Framework

AI Ready – Analysis Towards a Standardized Readiness Framework

hamper traditional disaster response efforts. To overcome these obstacles, a coordinated UAV
network is designed to autonomously perform essential tasks within disaster-stricken areas.
Utilizing RL algorithms, UAVs can learn and adjust their operations (including route navigation,
returning to charging stations, and data detection and transmission) based on feedback from
the environment. In particular, the project integrates several state-of-the-art RL approaches, such
as multiagent learning (for achieving efficient cooperation among UAVs), sim2real transfer (for
leveraging simulated data), and curriculum learning (for achieving a smoother learning curve
from simple to complex scenarios). This combination of approaches allows for the optimization
of task distribution and resource management in real-time, while ensuring generalization across
a rich variety of disaster scenarios.

UAVs will be equipped with sensors, cameras, communication systems, and payload delivery
mechanisms. The drones collaborate to carry out a range of tasks such as reconnaissance,
damage assessment, communication relay, and aid distribution. Through advanced data
collection and mapping algorithms, the UAV network achieves real-time situational awareness,
facilitating informed decision-making by the response teams.

To achieve the tasks, each UAV maintains a connection to ground stations, either through direct
links or an ad-hoc network, ensuring seamless coordination and data exchange.

The Ad-hoc networks used in the use case greatly facilitate communication among drones due
to their advantage in flexibility, mobility, cost-effectiveness, resilience, and scalability.

Since a disaster might destroy the existing infrastructure, the flexibility provided by ad-hoc
networks makes them ideal for temporary or emergencies where establishing traditional
networks is impossible. Devices in an ad-hoc network can move freely without affecting the
network's overall functionality, which makes the use of ad-hoc networks ideal for disaster
detection. Since ad-hoc networks don't require dedicated infrastructure, they can be more
cost-effective to deploy. Ad-hoc networks can also be more resilient compared to traditional
networks, as they don't rely on a central point of control. Ad-hoc networks can easily scale
to accommodate many devices, making it easy to scale to a large number of drones in real
applications.

10