Optimization of Obstacle Detection for Small UAVs

June 24, 2022

The UAV market grows every year and in order to keep this, it is required a large investment in Sense and Avoid systems. With this in mind, the thesis is centred on this type of systems with special focus on the detection phase in small fixed-wing UAVs. Firstly, a presentation of the various types of available sensors and avoidance algorithms is made culminating with a list of the most suitable sensors for this type of work and a background explanation on the Potential fields method that will be used as the avoidance algorithm. Next, parameters are defined to characterize the sensing systems and then tested in simulated obstacle avoidance missions. At the end of these simulations, several conclusions were drawn on the influence of the parameters in the overall performance of the system. The work continues with a presentation of the hardware and software required to implement a Sense and Avoid system. After the sensors and controller have been chosen, a schematic is drawn up with all the necessary devices and the connections between them. Furthermore, the flight controller firmware and the ground control station software are chosen taking into account their specifications. Next, the work proceeds to the elaboration of experiments with the sensors, obtaining real information about their capabilities such as maximum range, average error of the measurements, etc. Finally, a simple sense and avoid system is implemented in a small rover in order to perform experiences that can evaluate the capabilities of the sense and avoid system.

Authors in the community:

• 

  P

  Pedro Miguel Silva Serrano

  ist186677

Supervisors of this institution:

• 

  A

  André Calado Marta

  ist31052

mechanical-engineering - Mechanical engineering

por - Portuguese

Instituto Superior Técnico