Advanced search
Start date

Probabilistic robust control of Balloon-Multicopter aerial vehicles

Grant number: 17/06877-2
Support type:Scholarships abroad - Research
Effective date (Start): January 08, 2018
Effective date (End): July 07, 2018
Field of knowledge:Engineering - Aerospace Engineering
Principal Investigator:Davi Antônio dos Santos
Grantee:Davi Antônio dos Santos
Host: Constantino Manuel Costa Lagoa
Home Institution: Divisão de Engenharia Mecânica (IEM). Instituto Tecnológico de Aeronáutica (ITA). Ministério da Defesa (Brasil). São José dos Campos , SP, Brazil
Local de pesquisa : Pennsylvania State University, United States  


In the last few years, the multirotor aerial vehicles (MAVs) have found many applications. In most cases, the operation could become more effective and efficient if the flight duration and payload capacity were extended. A very simple way to improve an MAV in these two aspects is by combining it with a helium balloon that provides a net aerostatic lift counteracting the total weight of the vehicle. This six-month research project will deal with the dynamic modeling and robust design of flight control laws for a hybrid unmanned aerial vehicle (UAV) consisting of an oblate spheroid helium balloon coupled with a multirotor airframe containing four vertical fixed rotors. A nonlinear dynamic model with six degrees of freedom (DOF) is derived for this balloon-quadcopter using the Newton-Euler approach. Among other efforts to which the MAVs are usually subject, the proposed model includes a restoration torque due to the displacement of the balloon's center of buoyancy (CB) above the vehicle's center of mass (CM), a random variation of the air and helium densities, a second-order vibration representing a flexible connection between the balloon and the quadcopter airframe, and the bounding of the resultant thrust and torque vectors within appropriate sets that ensure the satisfaction of design limits for the rotors.The most common flight control framework in aerospace systems is adopted in this work. It is an hierarchical architecture, based on the so-called time-scale separation assumption, in which the attitude control is realized by a faster inner loop while the position control is carried out by a slower outer loop. This assumption allows to separately design the attitude and position control laws. In particular, in order to deal with the aforementioned control problem, we will investigate a probabilistic robust control design method based on a multi-input chattering-free optimal-sliding-mode strategy together with scenario optimization. The proposed methods will be evaluated by extensive Monte Carlo simulations.