Distributed Fault-Tolerant Containment Control for Linear Heterogeneous Multiagent Systems: A Hierarchical Design...

Area of Science:

• Control Systems Engineering
• Distributed Systems
• Robotics

Background:

• Existing distributed control methods for multiagent systems (MASs) suffer from fault propagation, where agent failures disrupt network dynamics.
• Agent failures in conventional MASs can cascade, affecting multiple agents and compromising overall system performance.

Purpose of the Study:

• To develop a distributed hierarchical fault-tolerant containment control protocol for heterogeneous linear multiagent systems.
• To prevent fault propagation in multiagent systems by isolating the impact of agent failures.

Main Methods:

• A two-layer hierarchical control architecture is proposed, comprising an upper layer for virtual containment and a lower layer for tracking.
• The hierarchical design isolates faults, ensuring that a failed agent's dynamics only affect itself, not the network.

Main Results:

• The developed fault-tolerant hierarchical protocol effectively avoids fault propagation across the network.
• Each follower agent asymptotically converges to a convex hull defined by the leaders, even with external inputs.

Conclusions:

• The hierarchical design is a viable approach to achieve fault tolerance in multiagent systems.
• This method enhances the robustness of heterogeneous linear multiagent systems against individual agent failures.