Design and Development of Drones to Autonomously Interact with Objects in Unstructured Outdoor Scenarios
Sirena Cascarano, Mario Milazzo, Andrea Vannini, Andrea Spezzaneve, Stefano Roccella
- Year
- 2021
- Citations
- 2
- Access
- Open access
Abstract
Aerial drones are systems that have been largely employed in a number of civil applications, from surveillance and environmental monitoring to day-to-day service. In view of this, the 2020 Mohamed Bin Zayed International Robotics Challenge (MBZIRC) held in Abu Dhabi (UAE) was a competition that aimed to push forward the technological limits of the actual aerial devices. We present the design and development of drone prototypes to meet one of the goals of Challenge 1, namely autonomously identifying and attacking randomly-positioned colored targets. This paper reports the main hardware/software design choices to accomplish the highest flexibility and usability of our aerial platforms, namely the strategies to control the overall dynamics and vision for the eventual interaction with targets. We achieved the fourth place among 22 international teams, since we successfully dispatched all the targets, which were dispersed throughout an area of 6000 m 2 , in 365 s. Although focused on specific goals, the methodologies developed in view of the competition are suitable for further improvements towards other application fields.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991