Development of autonomous quadcopter sensor platform for 2D localization and mapping

Increased usage of Micro Aerial Vehicles in everyday life has made autonomous quadrotor gain a lot of popularity in both robotics research and society. One key element of autonomous navigation is localization. Typical outdoor autonomous quadrotor localizes itself using GPS signal. However, flying indoor may impose a challenge as GPS may not be available or the localization is not detail enough. This project aim to solve this problem by using LIDAR laser scanner as the main sensor to perform SLAM (Simultaneous Localization and Mapping), which not only able to localize but also provide the map of the environment. Two SLAM algorithms (Hector SLAM and Gmapping with Canonical Scan Matcher) will be tested on semi-autonomous flight and compared side by side with Vicon motion capture system. The resulting localization data will then be used to estimate the corrected pose and velocity using state observers algorithm. Another important aspect of autonomous quadrotor, which is the position control of the quadrotor, will also be briefly analyzed. Position controller of the quadrotor will use the filtered pose and velocity to execute a setpoint-guided flight.Bachelor of Engineering (Mechanical Engineering