Finite-Time Robust Distributed Estimate for Nonlinear Systems With Heterogeneous Sensors

Area of Science:

• Control Systems Engineering
• Signal Processing
• Nonlinear Dynamics

Background:

• Distributed state estimation is crucial for networked systems with multiple sensors.
• Handling heterogeneous sensors in nonlinear systems presents significant challenges.
• Existing methods often struggle with finite-time convergence and data fusion.

Purpose of the Study:

• To develop a finite-time distributed state estimation (DSE) algorithm for discrete-time stochastic nonlinear systems.
• To address the complexities introduced by heterogeneous sensors in a network.
• To ensure accurate and robust state estimation within a finite time frame.

Main Methods:

• A three-phase framework: priori prediction, measurement update, and consensus fusion.
• Interactive Multiple Model (IMM) for accurate priori state prediction.
• A novel heterogeneous measurement information fusion algorithm using a measurement probability matrix.
• Consensus-based fusion with consensus weights for distributed state estimates.

Main Results:

• The proposed DSE algorithm achieves finite-time convergence.
• Demonstrated accurate state estimation for nonlinear systems with heterogeneous sensors.
• The algorithm's performance was validated through simulation examples.

Conclusions:

• The developed finite-time DSE algorithm effectively handles nonlinear systems with heterogeneous sensors.
• The three-phase framework ensures accurate state estimation and finite-time convergence.
• The novel fusion algorithm enhances the utilization of diverse sensor measurements.