Design of zonotopic space interpolation-based smooth variable structure filter for systems with unknown but bounded...

Area of Science:

• Control Systems Engineering
• Estimation Theory
• Robotics

Background:

• State estimation is crucial for dynamic systems.
• Unknown and bounded noise poses significant challenges.
• Conventional filters often suffer from conservativeness and chattering.

Purpose of the Study:

• To propose a novel Zonotopic Space Interpolation-based Smooth Variable Structure Filter (ZSI-SVSF).
• To address state estimation challenges under unknown but bounded noise.
• To improve filter efficiency and reduce conservativeness.

Main Methods:

• Combines zonotopic contraction for state confidence envelopes.
• Utilizes cubic spline interpolation for smooth boundary layers.
• Employs a curvature-driven mechanism for adaptive gain computation and iterative updates.

Main Results:

• ZSI-SVSF avoids covariance computations, enhancing efficiency.
• The filter mitigates chattering and reduces conservativeness.
• Demonstrated improved accuracy and adaptability in linear and nonlinear systems (UAV tracking).

Conclusions:

• ZSI-SVSF offers a robust and efficient solution for state estimation.
• The method is applicable to both linear and nonlinear systems.
• Validated for improved performance in complex dynamic scenarios.