Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
We introduce a three step procedure to improve attitude determination by MEMS-IMU sensors on board a small UAV. IMU pre-calibration is performed first and only once in the lab; second in a simplified version and for every switch-on in the field. Together, they remove the prevailing part of sensor deterministic errors, which in turn improves the quality of coarse and subsequently fine alignment. At the same time this allows to proceed further during a flight with noise reduction using multiple IMUs through a simple procedure, which is the third and final phase prior the conventional INS/GNSS integration via Kalman filtering and smoothing. The effectiveness of this approach are demonstrated practically i) with respect to a continuous absolute reference on board of a helicopter performing laser-scanning and ii) with a small fixed-wing UAV employed for high-resolution mapping.
Luis Guillermo Villanueva Torrijo, Damien Maillard, Göran Stemme, Simone Pagliano
Kyojin Choo, Li Xu, Yimai Peng