In this paper, the authors propose a cooperative formation control strategy with collision-avoidance capability for a multi-unmanned aerial vehicle (UAV) system using decentralized model predictive control (MPC) and consensusbased control. Consensus-based control algorithms are applied for formation flying in three-dimensional space. However, UAVs where these formation control algorithms are applied have not the ability to avoid collisions. Decentralized model predictive control (MPC) is applied to generate control inputs for formation flying with collision-avoidance capability. Using decentralized MPC, each UAV plans only its own action to track the trajectory specified by the formation control algorithm within the feasible regions satisfying collision-avoidance. The authors show how the optimization problems with coupled constraints such as collision-avoidance can be solved by each decoupled UAV in parallel with the other UAVs so that the decisions independently taken by each UAV can ensure consistency in coupled constraints of collisionavoidance. The computation time is also taken into account because it is a crucial factor to apply MPC to actual UAVs. Finally, the proposed approach is validated by some simulations.
: This paper, studies cooperative control problems with a multi-UAV (Unmanned Aerial Vehicle) system expressed as a fourth-order system using a consensus-based algorithm. How to model a linearized model of UAVs like quadrotors as a fourth-order system is described, and then a formation control algorithm for the fourth-order system is proposed after formulating a problem. The proposed control law is based on a consensus algorithm, and a leader-follower structure is also applied to the control law so that the leader can provide the quadrotors with commands such as their desired states. And then, The study shows that the proposed control algorithm can guarantee accurate formation keeping when fundamental assumptions about the network composed of the multiple UAVs are satisfied. Finally, the proposed approach is validated by some simulations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.