Publication: Obstacle alert and collision avoidance system development for UAVs with Pixhawk flight controller
Loading...
Identifiers
Publication date
2018-06
Defense date
2018-07-09
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
In recent years, the unmanned aerial vehicles sector has been characterized by its
sharp growth, spreading its line of applications and becoming one of the cutting
edge technologies in the world. However, this exponential advancement would have
been even more extreme but for the restrictive existing legislation that limits its
operations.
That constraints imposed to drone operations are not legislated in vain. Specially in
populated areas, flying drones is a dangerous service that entails risks for the safety
of the population. Useless have been the attempts of many major online selling
companies to make use of unmanned aerial vehicles serving as dealers of parcels.
Further technology needs still to be implemented, for guarantying standards levels
of safety in urban zones.
This thesis aims to contribute to this required development of drone technology
by proposing a preliminary collision avoidance system for unmanned aerial vehi-
cles. The project involves assembling a flight-capable quadcopter from scratch and
implementing the collision avoidance as an additional subsystem. To that end, a
set of ultrasonic range finders are located around the quadcopter. Their acquired
raw data is processed in an auxiliary arduino microcontroller board that send tra-
jectory corrections to the flight controller of the quadcopter based on the distance
information.
As a result, a concept proof of an autonomous collision avoidance system is integrated
into an unmanned aerial vehicle system. Results are obtained and consecutively
analyzed based on ground and flight tests. For that purpose, the
flight capabilities
of the built quadcopter are proved, and aftewards, the collision avoidance is tested.
Overall, the collision avoidance system e ectiveness was achieved. The collision
avoidance work principle consisted on warding off from obstacles when detected.
Major faced problems are related to stability recover after the collision is avoided.
That problem was solved by proving different flight modes, but that issue needs
future work to make the collision avoidance more reliable.
Description
Keywords
UAVs, Collision avoidance, Ultrasonic range finder, Arduino board, Pixhawk flight controller, Obstacle detection, Environment monitoring