Estimation and Control for Cooperative Slung Load Transportation by a Team of Multirotors
Work towards MSc degree under the supervision of Assoc. Prof. M. Idan (Technion)
Department of Aerospace Engineering
Technion – Israel Institute of Technology
Multirotor UAVs are becoming widespread for both civil and military uses, such as photography, surveillance and payload transportation. A slung load system, in which the payload is connected to several multirotors through cables, can be used as a modular payload transportation system, allowing significantly heavier and larger payloads to be carried, which cannot be transported using a single multirotor. Control and coordination of such multirotor-payload systems is address in this study.
The current research adopts a geometric control approach, in which the nonlinear and coupled dynamics of the multirotors, cables and payload can be explicitly incorporated into control system design. Using three or more multirotors, this approach enables a full 6DOF (pose) tracking of the payload. An on-line optimization scheme is used to enable safe and efficient transport maneuver, in terms of collision avoidance between multirotors and efficient weight distribution.
To cope with high levels of parameter uncertainty, which is bound to exist in a system of this kind, an on-line parameter identification scheme is used to identify the payload parameters, namely its mass, moments of inertia, and cable connection points. It is shown that system parameter identification and thus the reduction in parameter uncertainty contributes greatly to the robustness and tracking performance of the system.
Payload transportation using four Pegasus 120 octorotors of Aeronautics Ltd. is examined in a full scale 3D simulation, demonstrating that the suggested approach provides excellent payload pose tracking, even when the system is exposed to significant disturbances, measurement noise and various uncertainties.
The talk will be given in Hebrew
Wed, 24-02-2021, 16:30 (Zoom Meeting)https://technion.zoom.us/j/95034990472