EKF Design for Online Trajectory Prediction of a Moving Object Detected Onboard of a UAV
Milena Faria Pinto, Fabrício O. Coelho, João Pedro Carvalho de Souza, Aurélio G. Melo, André Luís Marques Marcato, Critina Urdiales · 2018
The applications with Unmanned Aerial Vehicles have increased in the last decades due to their economic and technical feasibility. Moreover, several tasks require online objects tracking as well as the object position knowledge in the real-world with algorithms execution onboard. An example of such task is the video surveillance with human activity recognition. In this paper, we propose a new approach using Extended Kalman Filter to estimate and to predict the object real-world coordinates. This research shows that the results were up to 30% better compared to the results without data processing.