Distributed zero-sum differential game for multi-agent systems in strict-feedback form with input saturation and...

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Robotics

Background:

• Investigates the distributed differential game tracking problem for nonlinear multi-agent systems.
• Addresses challenges including output constraints, uncertain nonlinearities, and input saturation in follower systems with strict-feedback structures.

Purpose of the Study:

• To develop a distributed control scheme for nonlinear multi-agent systems that ensures tracking performance while satisfying output constraints.
• To minimize the cooperative cost function in a zero-sum differential game context.

Main Methods:

• Utilizes command filtered backstepping and neural networks (NNs) to handle unknown nonlinearities and input saturation.
• Introduces a novel barrier Lyapunov function (BLF) to manage output constraints.
• Employs adaptive dynamic programming (ADP) with a critic network for online cost function approximation and strategy derivation.

Main Results:

• Guarantees that closed-loop signals are cooperatively uniformly ultimately bounded (CUUB).
• Ensures minimization of the cooperative cost function.
• Demonstrates that output constraints and input saturation are not violated.

Conclusions:

• The proposed distributed control scheme effectively solves the tracking problem for nonlinear multi-agent systems.
• The integration of BLF and ADP provides a robust solution for systems with complex constraints.
• Simulation results validate the efficacy of the developed control strategy.