Title:Nonlinear Model Predictive Control for 3D Formation of Multirotor Micro Aerial Vehicles with Relative Sensing in Local Coordinates

Abstract:The complex tasks such as surveillance, construction, search and rescue can benefit of the maneuverability of multirotor Micro Aerial Vehicles (MAVs) to obtain robust, cooperative system behavior and formation control is a prominent component of the these complex tasks. This work focuses on the problem of three-dimensional formation control of multirotor MAVs by using exclusively relative sensory information. It proposes a centralized Nonlinear Model Predictive Control (NMPC) approach in a leader-follower scheme. A realistic six degrees of freedom mathematical model of a multirotor MAVs is introduced and leveraged in the control laws. The formulation of the problem is performed based on NMPC and relative sensing framework with respect to local coordinate frames of the robots. This type of formulation makes the formation independent of the full knowledge of global or common reference frames and the utilization of expensive global localization sensors. Real-time Iteration (RTI) based solution to optimal control problem (OCP) is proposed by taking the novel formulation into account. An extensive scenario is designed to test and validate the strategy. Evaluation of the results suggests that satisfactory robust performance is achieved and maintained under model uncertainty and noise in local sensors and even in cases where the dynamics of the formation suddenly changes.