Jiang, Liwei (University of Stuttgart), Gonzalez, Luis Felipe (Queensland University of Technology (QUT)/ QUT Centre for Roboti), Mcfadyen, Aaron (Queensland University of Technology)

Cooperative Game Theory Based Multi-UAV Consensus-Based Formation Control

Scheduled for presentation during the Regular Session "Swarms" (WeA3), Wednesday, September 2, 2020, 10:20−10:40, 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 19, 2024

Abstract

This paper presents a cooperative differential game theory approach to designing a consensus-based formation control strategy for multi-UAVs(agents). The consensus-based formation control problem is defined as tracking task where the states of each agent are required to follow a given reference. Position states are given as fixed offsets between each agent to achieve a formation whilst velocity and attitude are should be identical amongst agents to archive consensus. Using a cooperative game theory based approach, each agent acts to reduce the weighted team cost to reach consensus. The control strategy design is based on open-loop design, closed-loop design, and a semi-open-loop design concept to better account for different hardware configurations. The control approach is validated via simulations to show how cooperative game theory based consensus formation control can outperform non-cooperative game theory control and general optimal control with respect to reducing formation deviation. The semi-open loop design is proven to be suitable for practical applications.