PSO Based Optimal Gain Scheduling Backstepping Flight Controller Design for a Transformable Quadrotor

Transformable Unmanned Aerial Systems (UASs) are increasingly attracting attention in recent years due to their maneuverability, agility and morphological capacities. They have overcame many limitations such as, multi-tasks problem, structural adaptation in flight, energy consumption, fault tolerant control, and maneuverability. Nevertheless, their variable geometries as well as the great number of actuators make them highly nonlinear and over-actuated systems, which are characterized by a slow transformation mechanism, variable mathematical models, and complex design and control architectures. In this article, we propose a simple and lightweight design of a transformable quadrotor, which allows to increase the geometric adaptability in flight, maneuverability, and speed of the transformation process by exploiting fast and performant servomotors. Since the Center of Gravity (CoG) of the quadrotor varies according to the desired shape, it results in a variation of the inertia and the control matrix instantly. These parameters play a crucial role in the system control and its stability, which is substantially a key difference compared to the classic quadrotor. Thus, a new generic model will be developed, which takes into account all these variations together and the asymmetry of the configurations. To validate the developed model, ensure the stability of our quadrotor, and improve the performance of the linear control strategies applied to these new drones, an optimal gain scheduling backstepping controller based on Particle Swarm Optimization (PSO) algorithm will be designed and tested. The realized prototype will be presented at the end of this work.