Multirotor Aerial Vehicles (MAVs) have become ubiquitous across various fields, ranging from civil engineering to agriculture. A key reason behind their widespread adoption may be attributed to their ability to perform tasks that other robotic systems cannot, such as aerial surveying, firefighting for large areas, search and rescue missions, large-scale structural inspections, and collaboration with human workers at height. Many of these applications require precise physical interaction with the surrounding environment, driving advancements in MAV design and control architectures to meet these demands. Conventional quadrotors can handle interaction tasks with superior energy efficiency when the task specifications do not require decoupled motions. However, when independent control of position and orientation is necessary, fully-actuated vehicles such as the TiltHex [1] and Omni-Morph [2] can realize such motion.
Despite this clear advantage over quadrotors, TiltHex cannot sustain its weight in all orientations, making it unsuitable for tasks in which maintaining constant arbitrary tool orientation throughout task execution is desired. These two possible solutions to this limitation are reported: employing an omnidirectional MAV, like the Omni-Morph, with a rigidly-attached link and equipping a fully-actuated MAV with an additional robotic arm. Compared to the first solution, the second approach increases cost and maintenance demands due to added mechanical complexity and additional DC motors. Consequently, the first solution is often preferred.
To enable interaction tasks, a rigid link is attached to the Omni-Morph platform. This vehicle is built to optimize energy usage without limiting the set of achievable maneuvers by switching between these two operating modes based on task requirements: performance mode for tasks that do not require decoupled motions like point-to-point trajectory following, and full motion dexterity mode when precise positioning and orientation are needed. The ultimate goal of this thesis is thus to obtain and demonstrate stable and reliable physical interaction operation with the Omni-Morph in real world experiments, with the performance remaining within satisfactory bounds. In particular, it aims to achieve the following outcomes:
- Evaluating the performance of an interaction controller with fixed propellers tilting angle —previously tested in ideal simulation conditions—through real-world experiments. These tests include sliding contact (with the tool having fixed and variable orientations), sliding on surfaces with different slopes, and peg-in-hole tasks, all benchmarked against established metrics.
- If time permits, extending the current control scheme to accommodate variable tilting angles and validating it through Gazebo simulations and experiments.
- Conducting a feasibility study of alternative control schemes and assessing their performance relative to the developed baseline controller.
Prerequisites
- Strong programming skills in MATLAB/Simulink.
- Solid understanding of fundamentals in robot dynamics and control.
- Proficiency in using realistic physics simulators like Gazebo.
References
[1] S. Rajappa, M. Ryll, H. H. Bülthoff, and A. Franchi, “Modeling, control and design optimization for a fully-actuated hexarotor aerial vehicle with tilted propellers,” in2015 IEEE International Conference on Robotics and Automation (ICRA), 2015, pp.4006–4013.
[2] Y. Aboudorra, C. Gabellieri, R. Brantjes, Q. Sablé, and A. Franchi, “Modelling, analysis, and control of omnimorph: an omnidirectional morphing multi-rotor uav,”Journal of Intelligent & Robotic Systems, vol. 110, no. 1, p. 21, 2024.2