These lectures notes are prepared to provide minimal fundamentals to address estimation and fusion applications in robotics. We tried to present these basis using the data fusion as a main frame. These notes are organized as follows:
• We first introduce the Bayesian Estimation as the main fundamental angle stone to deal with minimal estimation issues,
• We then develop Dynamic Estimation in order to cope with real estimation issues robots have to face usually. In the section we’ll talk about, among others, Kalman filtering.
• The fusion itself is then introduced. Notice, because of the lack of time / place, we don’t address here important issues such as the decision or the tracking topics. In particular we’ll address the centralized and un-centralized fusion and the data un-synchronization issues.
• Since nowadays, computer vision address very complex fusion issues, in particular the Visual SLAM, we just focus on the graphical models as tools to deal with complex systems such as SLAM’s. In this part we’ll show briefly Bayes Networks and the Factors Graphs.
• Finally A short Appendix gives few minimum reminders required to well understand these notes.