Abstract
Background: The aim of this paper is to investigate data fusion techniques based on an Extended Kalman Filter (EKF), and more specifically, the nonlinear dynamic estimation of a wheelchair navigation system.
Methods: Three data fusion techniques are presented and a comparison between them is studied. It combines the noisy measurement data coming from several sensors to obtain the best estimate of position while reducing the measurement uncertainties.
Results: By using the MATLAB, the performance of these techniques is checked with simulated data and performance metrics are calculated for evaluation of the algorithms. Detailed mathematical expressions are provided which could be useful for algorithm implementation.
Conclusion: The results show that the algorithm based on a measurement fusion technique gives a good estimate when compared with another one.