Comparison of Control Methods for Trajectory Tracking in Fully Actuated Unmanned Aerial Vehicles

Abstract

In this paper recently proposed control methods, suitable for trajectory tracking in fully actuated Unmanned Aerial Vehicles (UAVs), are reviewed and experimentally compared. We specifically focus on multirotor platforms with tiltable propellers, for which the thrust and torque delivered by each rotor can be oriented within the aircraft frame by means of servo-actuators. Firstly, the main assumptions underlying the control design model usually employed in the literature are discussed, pointing out the main limitations. Then, some recent control laws that have been proposed to address the trajectory tracking problem for these platforms are reviewed, placing emphasis on their tracking capabilities as well as on the ease of tuning and implementation. Finally, experimental results obtained by applying three control laws on a tilt-arm quadrotor UAV are shown to evaluate and compare their performance. Two experiments, representative of operating conditions for thrustvectoring UAVs, have been performed: a set-point tracking with level attitude and a full-pose trajectory tracking task.

Flatness-based Quadcopter Trajectory Planning and Tracking with Continuous-time Safety Guarantees

“…Since a 1 , a 2 > 0, we have q − q + a 1 ( q − q) + a 2 (q − q) = 2a 2 δ > 0. Therefore, condition (15) in Theorem 2 of [34] is satisfied, which implies that h q , h q are control sharing barrier functions. It is easy to see that functions h x , h x , h y , h y , h z , h z are all control sharing barrier functions, which means that (CBF-QP) is feasible for all (t, z)…”

“…Applying P2) of B-splines shown in Section II-B, we determine that s (r) (t) is a convex combination of r-th order VCPs, which is equivalent to (15). Using P3) of B-splines we have…”

“…Recently, there are also many works that aim to improve the tracking performance of quadcopters, particularly for aggressive flights [15], [16], [17], [18]. For example, [16] tracks highorder derivatives to reject aerodynamic effects present in aggressive flights, [17] considers a state-dependent linearization of the quadcopter dynamics and applies LQR techniques when the system is operated far from the equilibrium point, and [18] employs flatness results to feedforward-linearize the MPC problem for optimal control.…”

Towards a new design with generic modeling and adaptive control of a transformable quadrotor

“…Tilt-rotor UAVs have been proposed in different research works (21,(51)(52)(53)(54)(55)(56) , and tilt-body UAVs were presented in Refs (16,57,58) , are another category of transformable UAVs that presents several advantages such as enhanced agility and precision, high endurance and maneuverability, adjustment of the thrust force by compensation in the event of a breakdown, horizontal and vertical displacement, elimination of the over-actuation problem and higher speed. However, the main disadvantage of tilting rotors is their poor hover performance caused by the small rotor diameters and large rotor mast.…”

“…All of these challenges led us to focus on the potential of transformable quadrotors. This class of drones is usually reported as unconventional quadrotors (8) , transformable multirotors (9) , foldable drones (10,11) , bio-inspired UAVs (12) , drones with extendable arms (13) , morphing geometry quadcopters (14) , tilt-arm quadrotors (15) , quadcopters with a tilting frame (16) , deformable drones (17) , agile quadrotors (18) , multilink aerial robots (19) , drones with adaptive morphology (5) , drones with changeable geometry (20) , tiltable-rotor aerial vehicles (21) , convertible drones (22) or modular drones (23) . Nowadays, the vision of researchers towards such drones and their applications has changed radically, they have opened a new interesting research field in the automation and flying robots community (24) .…”