Adaptive unscented information filter for multiple sensor fusion

An adaptive nonlinear information filter for multiple sensor fusion is proposed in situations with nonlinear signal models and unknown process noise covariance. The proposed derivative free information filter, based on Unscented Transformation, ensures satisfactory estimation performance by online adaptation of the unknown process noise covariance. Efficacy of the proposed estimator is demonstrated with a well known object tracking problem in a sensor fusion configuration. From the results of Monte Carlo simulation it has been revealed that in the situation when the prior knowledge of the process noise covariance is lacking the performance of the proposed filter is demonstrably superior to its non adaptive version in the context of joint estimation of parameters and states.