Improving Interval Kalman Filtering Through Interval Optimization Strategy

This paper proposes a novel interval Kalman filtering algorithm that extends the standard interval Kalman filtering to wireless sensor networks, and incorporates global optimization techniques to search for the optimal estimation interval of interval systems containing uncertainties, thus enhancing the precision of the algorithm. Firstly, the classic evolutionary programming is incorporated into the improved upper bounded interval Kalman filter based on the ideas presented in evolutionary programming Kalman filter, in order to validate the feasibility of the proposed approach. Subsequently, the evolution strategy-interval optimization algorithm which is enhancive interval optimization strategy is employed to replace the evolutionary programming method for reduce optimization costs and enhance optimization effectiveness. Finally, a computational simulation example is provided and compared with relevant algorithms. The results demonstrate that the newly proposed method in this paper achieves higher accuracy and exhibits less conservatism compared to the alternatives.