UAV autonomous navigation by image processing and a new formulation for particle filter

Unmanned Aerial Vehicles (UAV) is a technology with strong development, with application on several fields such as engineering projects, agriculture, livestock, monitoring, and rescue, for citing few examples. For the autonomous navigation, a standard scheme is to use Global Navigation System by Satellite (GNSS), onboard. However such signal can suffer natural or human interference.
Our approach applies image processing procedure: image edge extraction with correlation matching with georeferencing image. A data fusion is also applied, for combining the inertial sensor estimation and imagem processing positioning schemes. The data fusion is performed by using a new particle filter, where the likelihood operator is derived from the Tsallis’ distribution for the generalized statistical mechanics. The methodology was successful in performing the data fusion, generating good results and allowing to compute a confidence interval.