Distributed Mode-Dependent Event-Triggered Passive Filtering for Flexible Manipulator with Semi-Markov Parameters

Area of Science:

• Robotics
• Control Systems Engineering
• Networked Systems

Background:

• Flexible robotic manipulators present challenges in state estimation due to their complex dynamics.
• Distributed filtering over sensor networks is crucial for real-time control and monitoring.
• Existing methods may not adequately address varying loads and unstructured environments.

Purpose of the Study:

• To investigate the passive filtering problem for flexible robotic manipulators in sensor networks.
• To develop a distributed filtering algorithm that ensures prescribed passive performance.
• To enhance communication efficiency using a novel event-triggering mechanism.

Main Methods:

• Utilizing a semi-Markov model for flexible manipulators with varying loads.
• Developing a mode-dependent event-triggering mechanism for distributed filter communications.
• Applying model transformation and convex optimization to design mode-dependent filters.

Main Results:

• Sufficient conditions for guaranteed passive performance under disturbances were established.
• Mode-dependent filters were successfully designed using convex optimization.
• Simulations demonstrated the effectiveness of the developed filtering algorithm for a single-link manipulator.

Conclusions:

• The proposed distributed filtering algorithm effectively addresses the passive filtering problem for flexible robotic manipulators.
• The semi-Markov model and mode-dependent event-triggering mechanism offer practical advantages for real-world applications.
• The developed method ensures robust state estimation and passive performance in challenging environments.