Adaptive Distributed Fusion Estimation for Multi-sensor Nonlinear System with Bounded Noise

For the state estimation problem of multi-sensor nonlinear systems with unknown parameters and unknown-but-bounded noise, a distributed fusion estimation is proposed. The method firstly converted the nonlinear system to a new linear system by Taylor series expansion. Then an adaptive set-membership filter (ASMF) is designed as the local optimal filter for each sensor system based on set membership estimation. Subsequently, a global optimal filter is obtained by using the union geometry method of the Minkowski sum. Finally, the simulation results verify the feasibility and effectiveness of the proposed adaptive distributed fusion estimation algorithm.