Predefined-Time Formation Control for Multiagent Systems-Based on Distributed Optimization.

Area of Science:

• Robotics
• Control Theory
• Distributed Systems

Background:

• Leaderless formation control is crucial for multiagent systems coordination.
• Existing methods often require complex parameter tuning or auxiliary variables.
• Minimizing global functions with local interactions presents significant challenges.

Purpose of the Study:

• To develop a distributed optimization strategy for leaderless formation control in first-order multiagent systems.
• To achieve global function minimization and coordinated formation within a predefined time.
• To propose a method with fewer adjustable parameters and broader applicability to complex cost functions.

Main Methods:

• A two-step distributed optimization process is proposed.
• Step 1: Local controllers guide each agent to its local function's minimizer.
• Step 2: Agents are guided towards leaderless formation and global function minimization.

Main Results:

• The proposed scheme requires fewer adjustable parameters compared to existing methods.
• It effectively handles highly nonlinear, multivalued strongly convex cost functions.
• Agents do not need to share gradients or Hessians, enabling decentralized operation.
• Simulations confirm the effectiveness and superiority over state-of-the-art algorithms.

Conclusions:

• The presented approach offers an effective and efficient solution for leaderless formation control.
• It simplifies control design by reducing parameter dependency and enabling use of complex cost functions.
• The method demonstrates robust performance and broad applicability in distributed multiagent systems.