This project utilises parcel delivery using MARL, where autonomous drone agents learn to collaborate efficiently to complete search and delivery tasks. Human supervisors will be able to issue high-level commands, while the MARL agents interpret, coordinate and execute parcel delivery in a city.

One Page Doc

UML

Flowchart

Fig 11. Flowchart for the environment and drone behaviour.

Programming Architecture

Fig 12. Programming Architecture for ML Agent, training logic.

Delivering essential supplies to unknown clients that are living off the grid - campers, survivalists or those recovering from disasters.

For each delivery task, a pair of drones; a search drone and a delivery drone, will be sent out together. The client initiates the process by sending out a ping. The search drone will scan the area to locate the client and then communicate the client's position to the delivery drone. Once the delivery drone receives this information, it picks up the supplies and delivers it to the client.

Both drones will collaborate and share the same rewards for a successful delivery.