Narrow Results

For aerial manipulators, a steady flight must be guaranteed to perform safe interaction with the surrounding environment. This paper focuses on the development of a position control algorithm for an aerial manipulator system (AMS). The position control algorithm is based on the Immersion and Invariance (I&I) theory. The proposed controller maintains the position of the aerial manipulator at the desired point under external and internal disturbances. The control architecture uses a Visual-SLAM technique using on-board sensors for AMS positioning. A series of outdoor experimental tests are performed to demonstrate the effectiveness of the proposed control strategy.

This paper presents the internal architecture of a Modifiable Off-the-Shelf (MOTS) open-source autopilot. The analysis of a set of functional and hardware requirements reveals that the core of the autopilot can be implemented as a single-threaded system, with a main loop acting as a non-preemptive static scheduler, provided that the internal structure is well-organized. We discuss how the type of bus used for sensor communication influences the nature of the events received from the sensors, whether they are solicited or unsolicited. We demonstrate that, depending on the workload that a main loop iteration must handle, the execution time of a single iteration can exceed the defined period, potentially causing delays in attitude correction. Finally, we explore the degrees of freedom available to mitigate the impact of such overloads by smoothing out the periodic workload.