Autonomous Landing of a UAV on a Moving Platform

Due to the high popularity that drones are gaining over the last decades, a lot of researchers are focusing their attention to solve problems related to autonomous flight. Between all the possible task that can be performed using drones, the autonomous landing of an unmanned aerial vehicle on a moving platform seems to be a very good solution to some of the problems that arise using drones, as for example the limited operational range. In this work, the autonomous landing on a moving platform is solved using a low-cost solution. In fact, the hardware platform used presents a limited computational power and is equipped with a moderate number of sensors. Thanks to the compatibility of the micro-UAV with Matlab and Simulink, a mobel-based software design approach has been implemented. Moreover, the code generation principle has been used to transform models in C++ code that can be run on the on board processor of the drone. Starting from a project developed by MathWorks, the flight control system of the flying robot has been designed. In particular, a vision system able to get information from the environment and a path planning algorithm have been developed taking into account all the limitations imposed by the hardware platform. Once that the simulated results match the expected ones, the control system has been tested in a real environment in order to validate the design. In addition to that, the performances of this implementation have been analyzed in different conditions and finally compared.