Scheduling of multi-agent systems is a topical problem in AI and robotics [1]. We propose an algorithm for efficient and effective information gathering by drones in a real topological environment. Various types of information can be gathered – traffic information, facial recognition, etc. The algorithm manages the agents (drones) to acquire the maximum amount of information, considering limited resources.

Deliverables:

  • An implementation of the algorithm
  • A simulation suit for testing the algorithm
  • A visualisation that combines data from the OpenStreetMap and the output of the simulation
  • A tool for converting OpenStreetMap road networks into topological networks
  • A RESTful API for 3rd party services

[1] Yu, Karaman, Rus – "Persistent Monitoring of Events with Stochastic Arrivals at Multiple Stations" http://arxiv.org/abs/1309.6041

Built With

Share this project:

Updates