Leong, Wai Lun (National University of Singapore), Martinel, Niki (University of Udine), Huang, Sunan (National Universtiy of Singapore), Micheloni, Christian (Department of Mathematics, Computer Science and Physics, Univers), Foresti, Gianluca (University of Udine), Teo, Rodney (Temasek Laboratories, National University of Singapore)

Integrated Perception and Tactical Behaviours in an Auto-Organizing Aerial Sensor Network

Scheduled for presentation during the Regular Session "Networked Swarms II" (WeC3), Wednesday, September 2, 2020, 17:20−17:40, Edessa

2020 International Conference on Unmanned Aircraft Systems (ICUAS), September 1-4, 2020 (Postponed from June 9-12, 2020), Athens, Greece

This information is tentative and subject to change. Compiled on April 25, 2024

Abstract

We investigated the feasibility and conducted a proof-of-concept demonstration of an auto-organizing sensor network composed of UAVs and ground cameras, for urban surveillance. The solution is based on a decentralised paradigm with tightly coupled perception and tactical behaviour algorithms.

The sensor network can perform self-organization. New cameras that could be added any time would assign themselves to tasks. The network would reconfigure when cameras are removed or lost to ensure that high priority tasks are always served. Tracked targets could also be handed over from one camera to another, to ensure continuous tracking. This was achieved by means of our target re-identification algorithm and handover logic.

To realise our network, we selected and improved the state-of-the-art algorithms. We developed the architecture and integrated the algorithms. Extensive simulations and tests were conducted and finally, we successfully ran real world proof-of-concept tests, demonstrating the feasibility of our tightly coupled perception and tactical behaviour autoorganizing aerial sensor network.