A MIXTURE-OF-EXPERTS FRAMEWORK FOR ADAPTIVE KALMAN FILTERING

This paper proposes a modular and flexible approach to adaptive Kalman filtering using the framework of a mixture-of-experts regulated by a gating network, Each expert is a Kalman filter modeled with a differen t realization of the unknown system parameters such as process and mea surement noise, The gating network performs on-line adaptation of the weights given to individual filter estimates based on performance, Thi s scheme compares very favorably with the classical Magill filter bank , which is based on a Bayesian technique, in terms of i) estimation ac curacy, ii) quicker response to changing environments, and iii) numeri cal stability and computational demands, The proposed filter bank is f urther enhanced by periodically using a search algorithm in a feedback loop, Two search algorithms are considered, The first algorithm uses a recursive quadratic programming approach which extremizes a modified maximum likelihood function to update the parameters of the best perf orming filter in the bank, This particular approach to parameter adapt ation allows a real-time implementation, The second algorithm uses a g enetic algorithm to search for the parameter vector and is suited for post-processed data type applications. The workings and power of the o verall filter bank and the suggested adaptation schemes are illustrate d by a number of examples.