Dynamic Control of the DLR Suspended Aerial Manipulator

MSc assignment

The DLR Suspended Aerial Manipulator (SAM) [1] is an octarotor platform equipped with a KUKA LWR robotic arm. It was developed in the scope of the AEROARMs project to perform inspection and maintenance tasks in high and inaccessible industrial sites [2].

The SAM is suspended by a mobile crane, which sustains most of its weight. This allows for the use of relatively smaller propellers, compared to an autonomous detached helicopter,  which guarantee the safety of the operation while still being able to perform heavy-load tasks. Such smaller propellers can stabilize the platform under its suspension point and counteract the forces exerted by the arm, however, they cannot fully compensate for additional gravity effects in case the platform moves sensibly away from its initial pose (like a pendulum that can hang without effort, but needs increasingly higher force to swing up). Therefore, traditional reactive control methods cannot be applied straighforwardly as they do not account for such actuation limitation. On the other side, by using Nonlinear Model Predictive Control (NMPC) tecniques the members of our team (in particular D. Bicego and A. Franchi) have achieved good results when taking similar constraints into account on different aerial systems [3]. 

The goal of the master thesis will be to study the real time NMPC methods introduced in [3] and extend them in order to design and implement a novel controller for the the SAM that will expand its manipulation workspace through dynamic swing up maneuvers, which do not assume a full gravity compensation by the propellers at all the time. The developed controller will have to be tested in a realistic simulator, and, if successful, will be implemented on the real SAM at DLR.

The student will work at RAM and will be co-supervised also by A. Coelho (PhD candidate advised by A. Franchi) who works at the German Aerospace Center (DLR), near Munich. A short stay at the DLR Institute for Robotics and Mechatronics in the suburb of Munich for the final experiments can be planned for the student, but is not mandatory.

If you are interested in the project or looking for additional information, please, please contact

  • Andre Coelho [a.f.coelho@utwente.nl]
  • with A. Franchi [a.franchi@utwente.nl] and D. Bicego [d.bicego@utwente.nl] in CC.  

When applying, please include

  • an updated Curriculum Vitae,
  • a presentation letter highlighting your strengths and weaknesses,
  • and an updated exam grade list.

We are looking forward to your application.

[1] Sarkisov, Yuri S., Min Jun Kim, Davide Bicego, Dzmitry Tsetserukou, Christian Ott, Antonio Franchi, and Konstantin Kondak. "Development of sam: cable-suspended aerial manipulator." In 2019 International Conference on Robotics and Automation (ICRA), pp. 5323-5329. IEEE, 2019.

[2] https://youtu.be/onOc05Ymxzs?t=6

[3] Bicego, Davide, Jacopo Mazzetto, Ruggero Carli, Marcello Farina, and Antonio Franchi. "Nonlinear model predictive control with enhanced actuator model for multi-rotor aerial vehicles with generic designs." Journal of Intelligent & Robotic Systems 100, no. 3 (2020): 1213-1247.