In this letter, we address a challenging version of the problem of controlling tethered aerial vehicles (also known as UAV, MAV, and UAS) by considering the aerial robot linked to a generic and independently moving platform. We solve the exact tracking control problem for both the 3-D position of the robot (either absolute or with respect to the platform) and the tension along the link. To achieve this goal, we prove some fundamental system properties, useful to design a nonlinear controller, such as differential flatness and dynamic feedback linearizability. To close the control loop, a set of minimal and standard sensors is proposed. Then, we show that it is possible to retrieve the full system state from those sensors by means of nonlinear measurements transformations and a bank of low-dimension estimators based on the nonlinear high gain observer. The ability of the proposed observer-controller method is tested by extensive numerical simulations spanning many nonideal conditions.
Observer-Based Control of Position and Tension for an Aerial Robot Tethered to a Moving Platform / Tognon, M.; Dash, S. S.; Franchi, A.. - In: IEEE ROBOTICS AND AUTOMATION LETTERS. - ISSN 2377-3766. - 1:2(2016), pp. 732-737. [10.1109/LRA.2016.2523599]
Observer-Based Control of Position and Tension for an Aerial Robot Tethered to a Moving Platform
Franchi A.
2016
Abstract
In this letter, we address a challenging version of the problem of controlling tethered aerial vehicles (also known as UAV, MAV, and UAS) by considering the aerial robot linked to a generic and independently moving platform. We solve the exact tracking control problem for both the 3-D position of the robot (either absolute or with respect to the platform) and the tension along the link. To achieve this goal, we prove some fundamental system properties, useful to design a nonlinear controller, such as differential flatness and dynamic feedback linearizability. To close the control loop, a set of minimal and standard sensors is proposed. Then, we show that it is possible to retrieve the full system state from those sensors by means of nonlinear measurements transformations and a bank of low-dimension estimators based on the nonlinear high gain observer. The ability of the proposed observer-controller method is tested by extensive numerical simulations spanning many nonideal conditions.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


